Quinoline carboxamides as modulators of Breast Cancer Resistance

Transcription

Quinoline carboxamides as modulators of Breast Cancer
Resistance Protein (ABCG2):
Investigations on potency, selectivity, mechanism of action,
cytotoxicity, stability and drug-like properties
Dissertation
zur Erlangung des Doktorgrades der Naturwissenschaften (Dr. rer. nat.)
der Fakultät für Chemie und Pharmazie
der Universität Regensburg
vorgelegt von
Stefanie Bauer
aus Passau
2014
Die vorliegende Arbeit entstand in der Zeit von Januar 2011 bis Februar 2014 unter der Anleitung von
Herrn Prof. Dr. Armin Buschauer und Herrn Prof. Dr. Günther Bernhardt am Institut für Pharmazie
der Naturwissenschaftlichen Fakultät für Chemie und Pharmazie der Universität Regensburg.
Das Promotionsgesuch wurde eingereicht im Februar 2014.
Tag der mündlichen Prüfung: 28.02.2014
Prüfungsausschuss:
Prof. Dr. J. Heilmann
(Vorsitzender)
Prof. Dr. A. Buschauer (Erstgutachter)
Prof. Dr. G. Bernhardt (Zweitgutachter)
Prof. Dr. J. Wegener
(Prüfer)
Für
meine Eltern
und
meinen Verlobten
Mario
So eine Arbeit wird eigentlich nie fertig,
man muss sie für fertig erklären, wenn man nach
Zeit und Umständen das Mögliche getan hat.
Johann Wolfgang von Goethe (1749 - 1832)
Danksagung
An dieser Stelle bedanke ich mich herzlich bei:
Herrn Prof. Dr. Armin Buschauer, der mir die Möglichkeit gegeben hat, an einem Thema zu arbeiten,
das spannend und vielseitig zugleich war, für seine wissenschaftlichen Anregungen und die
konstruktive Kritik bei der Durchsicht dieser Arbeit.
Herrn Prof. Dr. Günther Bernhardt für seine umfassende Unterstützung im Labor, für seine
wissenschaftlichen Anleitungen und sein enormes Wissen sowie für sein stetes Interesse am
Fortgang der Experimente und für die konstruktive Kritik bei der Durchsicht dieser Arbeit.
Herrn Dr. Thilo Spruß für die Betreuung bei der Durchführung der tierexperimentellen Arbeiten sowie
der histologischen Untersuchungen.
Herrn Franz Wiesenmayer für die hervorragende Zusammenarbeit bei den tierexperimentellen
Arbeiten. Ich hoffe, Du genießt Deinen wohlverdienten Ruhestand.
Frau Petra Pistor für die Anfertigung der histologischen Präparate und Färbungen.
Herrn Prof. Dr. Burkhard König, Herrn Dr. Cristian Ochoa-Puentes, Herrn Manuel Bause sowie Frau
Qiu Sun für die Synthese der ABC-Transporter-Modulatoren und die gute Zusammenarbeit.
Frau Simone Stark für die gute Zusammenarbeit im Rahmen ihrer Masterarbeit.
Herrn Prof. Dr. H.-J. Galla und Frau Sabine Hüwel (Universität Münster) für die Testungen am
Bluthirnschranke-Modell.
Herrn Dr. Michael Thormann und Frau Laure Bourbon (Origenis GmbH, Martinsried) für die
Messungen zur Proteinbindung.
Frau Maria-Beer-Krön für die hervorragende Unterstützung bei der Durchführung verschiedenster
Versuche sowie ihre grenzenlose Hilfsbereitschaft im Laboralltag, ihre ansteckende Herzlichkeit und
die netten Überraschungen auf unseren Schreibtischen.
Frau Dita Fritsch und Frau Elvira Schreiber für die finalen Wiederholungen der ABC Transporter
Testungen.
Frau Brigitte Wenzl für die Praktikumsvorbereitung und die Durchführung der Mycoplasmen-Tests.
Herrn Peter Richthammer für seine stete Hilfsbereitschaft und Kompetenz bei allen technischen
Herausforderungen.
Herrn Johannes Felixberger für seine geduldige Hilfe bei den Klonierungsexperimenten.
Herrn Paul Baumeister, für die gemeinsame Teilnahme am Graduiertenprogramm der Emil Fischer
Graduate School in Erlangen sowie für die gegenseitigen „Sprechstunden“, Kaffeepausen und die
schöne Zeit im Labor.
all meinen derzeitigen und ehemaligen Kollegen, ganz besonders bei Miriam Ertel, Stefan Huber,
Nicole Kagermeier, Carolin Meyer, Nikola Pluym sowie meinem besten Freund und Kollegen Steffen
Pockes für die schöne Zeit mit ihnen in Regensburg.
allen meinen Praktikanten und Forschungspraktikanten insbesondere Christian Aigner und Vanessa
Niebauer.
sowie allen weiteren Mitgliedern des Lehrstuhls für ihre Kollegialität und Hilfsbereitschaft und das
angenehme Arbeitsklima in unserer Arbeitsgruppe.
Des Weiteren geht mein besonderer Dank an:
Meine Kollegen in der ehemaligen Engel-Apotheke in Barbing, insbesondere an Martina, mit der die
Samstage wie im Flug vergingen.
Alle Freunde außerhalb der Universität, im Besonderen an Eva für die ereignisreichen Urlaube und
Abende, an Birgit, für ihre aufmunternden Worte und ihre stets passenden Ratschläge sowie an Liesi,
für ihre gute Laune und ihr offenes Ohr.
Meine Familie, allen voran meinen Eltern, die mich von Beginn an unterstützt haben und deren Hilfe
ich mir zu jeder Zeit sicher sein konnte.
Meinen Verlobten Mario, der mir vor Augen geführt hat, worauf es im Leben ankommt - für seine
bedingungslose Liebe, sein Verständnis und seine Unterstützung.
Publications, Oral Presentations and Posters
Prior to submission of this thesis, results were published in part.
Publications:
C. Ochoa Puentes, P. Höcherl, M. Kühnle, S. Bauer, K. Bürger, G. Bernhardt , A. Buschauer, B. König
(2011); Solid phase synthesis of tariquidar-related modulators of ABC transporters preferring breast
cancer resistance protein. Bioorg. Med. Chem. Lett. 21 (12), 3654–3657.
C. Ochoa Puentes and S. Bauer, M. Kühnle, G. Bernhardt , A. Buschauer, B. König (2013);
Benzanilide – Biphenyl Replacement: A Bioisosteric Approach to Quinoline Carboxamide-type ABCG2
Modulators. ACS Med. Chem. Lett. 4 (4), 393-396.
S. Bauer, C. Ochoa Puentes, Q. Sun, M. Bause, G. Bernhardt , B. König, A. Buschauer (2013);
Quinoline Carboxamide-Type ABCG2 Modulators: Indole and Quinoline Moieties as Anilide
Replacements. ChemMedChem 8, 1773-1778.
K. Pollinger, R. Hennig, S. Bauer, M. Breunig, J. Tessmar, A. Buschauer, R. Witzgall, A. Göpferich
(2014); Biodistribution of Quantum Dots in the Kidney After Intravenous Injection. J. Nanosci.
Nanotechnol. 14, 3313-3319.
Short lectures:
Joint Meeting of the Austrian and German Pharmaceutical Societies, University of Innsbruck,
September 20-23, 2011
Overcoming ABCG2-mediated Drug Resistance with New ABCG2 Modulators Derived from Tariquidar
Bad Herrenalber Transporter- und Barriere-Tage, Bad Herrenalb, May 14-16, 2012
New Highly Potent and Selective Modulators of BCRP
Research Day of the Emil Fisher Graduate School of Pharmaceutical Sciences and Molecular
Medicine, Institute of Pharmacy and Food Chemistry, University of Erlangen, July 7, 2013
Inhibition of the Efflux Transporter ABCG2: A Strategy to overcome the Blood-Brain Barrier and
Chemoresistance
Poster presentations:
XXIInd International Symposium on Medicinal Chemistry, Berlin, September 2-6, 2012
and
Research Day of the Emil Fischer Graduate School of Pharmaceutical Sciences and Molecular
Medicine, Institute of Pharmacy and Food Chemistry, University of Erlangen, July 19, 2012
N-Acylated anthranilic acid esters: Highly potent and selective modulators of Breast Cancer
Resistance Protein (BCRP, ABCG2)
6th Summer School Medicinal Chemistry, University of Regensburg, September 26-28, 2012
Optimization of highly potent and selective ABCG2 transporter modulators with respect to solubility
and chemical stability
Professional training:
04/2011
Fortbildung für Projektleiter und Beauftragte für Biologische Sicherheit
(§15 und 17 Gentechniksicherheitsverordnung).
University of Regensburg, Germany
03/2012
Radioanalytik - Umgang mit offenen radioaktiven Stoffen.
03/2012 - 07/2012
Weiterbildung „Versuchstierkunde und Tierschutz“ (Bestandteil des
Nachweises der Sachkunde für den Umgang mit Versuchstieren innerhalb der
EU, FELASA Kategorie B)
06/2012 – 01/2014
Member of the Emil Fischer Graduate Programme
University of Regensburg & University of Erlangen, Germany
Contents
IX
Contents
1
General introduction .................................................................................................. 1
1.1
The ABC protein superfamily ................................................................................................... 1
1.2
Physiological and pharmacological functions of ABC transporters ......................................... 2
1.2.1
Structures and cellular mechanisms of ABC transporters ............................................... 2
1.2.2
ABC transporters and the phenomenon of multidrug resistance ................................... 4
1.2.3
ABC Transporters expressed at the blood-brain barrier ................................................. 6
1.3
2
Scope and Objectives ............................................................................................... 11
2.1
3
References ............................................................................................................................... 8
References ............................................................................................................................. 12
Modulation of the Breast Cancer Resistance Protein (ABCG2) ................................... 13
3.1
Introduction ........................................................................................................................... 13
3.1.1
The ABCG2 transporter ................................................................................................. 13
3.1.2
BCRP inhibitors .............................................................................................................. 14
3.2
Objective................................................................................................................................ 15
3.3
Materials and methods ......................................................................................................... 15
3.3.1
Drugs and chemicals ...................................................................................................... 15
3.3.2
Test compounds ............................................................................................................ 16
3.3.3
New fluorescent ABCG2 modulators ............................................................................. 24
3.3.4
Cell culture..................................................................................................................... 24
3.3.5
Cell based assays for the determination of ABC transporter modulation .................... 25
3.3.5.1
ABCG2 modulation: Hoechst 33342 and pheophorbide a microplate assays........... 25
3.3.5.2
Mitoxantrone microplate assay for the determination of ABCG2 modulation ........ 27
3.3.5.3
Calcein-AM efflux assay for the determination of ABCB1 modulation ..................... 28
3.3.5.4
Calcein-AM efflux assay for the determination of ABCC1 modulation ..................... 28
3.3.6
Chemosensitivity assays ................................................................................................ 29
3.3.7
Chemical stability of ABCG2 modulators in human and mouse plasma ....................... 29
3.3.7.1
Determination of the activity of unspecific esterases............................................... 30
3.3.7.2
Assay procedure ........................................................................................................ 30
3.3.7.3
HPLC and HPLC-MS analysis ...................................................................................... 31
3.3.8
3.4
ABC transporter modulation in a blood-brain barrier model........................................ 31
Results and discussion ........................................................................................................... 33
3.4.1
Inhibition of ABCB1, ABCC1 and ABCG2 ........................................................................ 33
3.4.2
Fluorescence properties of selected ABCG2 modulators .............................................. 39
X
Contents
3.4.3
Alternative fluorescence based ABCG2 inhibition assays ............................................. 40
3.4.4
Effect of co-administration of ABCG2 inhibitors with topotecan on the proliferation ....
of MCF-7/Topo cells ...................................................................................................... 44
3.4.5
Chemical stability of selected ABCG2 transporter modulators under .............................
physiological conditions ................................................................................................ 48
3.4.5.1
Chemical stability in culture medium and human CPD plasma................................. 49
3.4.5.2
Chemical stability in mouse plasma .......................................................................... 50
3.4.5.3
Esterase activity in human and murine plasma ........................................................ 55
3.4.6
4
Effect of selected ABCG2 transporter modulators on the transport of Daunorubicin ....
in a blood-brain barrier model ...................................................................................... 55
3.5
Summary and conclusions ..................................................................................................... 58
3.6
References ............................................................................................................................. 60
Drug-like properties of new ABCG2 modulators ........................................................ 65
4.1
Introduction ........................................................................................................................... 65
4.1.1
Trends in medicinal chemistry and drug development ................................................. 65
4.1.2
The ‘Rule of Five’ ........................................................................................................... 65
4.1.3
Plasma protein binding.................................................................................................. 66
4.2
4.2.1
4.2.2
Solubility ........................................................................................................................ 68
4.2.2.1
Comparison of solubilities in different media ........................................................... 68
4.2.2.2
Fluorescence spectra of Hoechst 33342 after intercalation in DNA in the ..................
absence and presence of test compounds ................................................................ 69
4.2.3
4.3
Plasma protein binding studies ..................................................................................... 69
4.2.3.1
Ultrafiltration ............................................................................................................. 69
4.2.3.2
Isothermal titration calorimetry ................................................................................ 70
4.2.3.3
Equilibrium dialysis .................................................................................................... 70
4.2.3.4
Protein binding analysis via fluorescent spectroscopy ............................................. 71
4.2.3.5
HPLC-based methods to determine protein binding of selected ABCG2 .....................
modulators ................................................................................................................ 72
Results and discussion ........................................................................................................... 73
4.3.1
Solubility of selected ABCG2 modulators ...................................................................... 73
4.3.1.1
Computational calculations ....................................................................................... 73
4.3.1.2
Solubility in DMSO and PBS ....................................................................................... 74
4.3.1.3
Fluorescence spectra of the Hoechst 33342 dye in the presence of ...........................
different modulators ................................................................................................. 75
4.3.2
Extent of plasma protein binding .................................................................................. 76
4.3.2.1
Ultrafiltration and ITC ................................................................................................ 76
4.3.2.2
Equilibrium dialysis .................................................................................................... 76
Contents
4.3.2.3
Investigations on fluorescent modulators to determine protein binding................. 77
4.3.2.4
Determination of lipophilicity parameters by HPLC .................................................. 82
4.3.3
5
XI
pH-dependent fluorescence of ABCG2 modulators ...................................................... 83
4.4
Summary and conclusions ..................................................................................................... 86
4.5
References ............................................................................................................................. 87
Characterization of human brain tumor cell lines...................................................... 91
5.1
Malignant brain tumors......................................................................................................... 91
5.1.1
Classification .................................................................................................................. 91
5.1.2
Incidence and mortality................................................................................................. 91
5.1.3
The blood-brain barrier and brain cancer therapy ........................................................ 92
5.2
5.2.1
5.2.2
Cell lines and culture conditions ................................................................................... 94
5.2.3
Cell staining ................................................................................................................... 95
5.2.4
Genetic stability - Karyology .......................................................................................... 95
5.2.5
Chemosensitivity against common cytostatic drugs and growth kinetics .................... 95
5.2.6
Tumorigenicity and growth kinetics of subcutaneous tumors in nude mice ................ 96
5.2.7
Histology ........................................................................................................................ 96
5.2.8
Induction of BCRP overexpression in brain tumor cells ................................................ 96
5.2.9
ABC transporter detection by Western Blot analysis .................................................... 97
5.2.9.1
Cell lysis and protein quantification .......................................................................... 97
5.2.9.2
SDS-PAGE and Western blot ...................................................................................... 97
5.2.10
5.3
ABC transporter detection by flow cytometry .............................................................. 98
Results and discussion ......................................................................................................... 100
5.3.1
Morphology ................................................................................................................. 100
5.3.2
In vitro growth of brain tumor cells............................................................................. 101
5.3.3
Aneuploidy................................................................................................................... 102
5.3.4
Chemosensitivity against cytostatic drugs .................................................................. 104
5.3.5
Characterization and growth kinetics of human brain tumor cell lines ...........................
in a subcutaneous tumor model in nude mice ............................................................ 112
5.3.5.1
In vivo growth kinetics ............................................................................................. 112
5.3.5.2
Histology .................................................................................................................. 113
5.3.6
Investigations on ABCG2 induced cell lines ................................................................. 113
5.3.6.1
Western Blot analysis of wildtype and induced cell lines ....................................... 113
5.3.6.2
Determination of ABCG2 overexpression by flow cytometry ................................. 115
5.3.6.3
Hoechst 33342 assay using ABCG2 induced cancer cells ........................................ 116
5.3.6.4
Chemosensitivity of LN-18/Topo cells against selected cytostatics ........................ 118
XII
Contents
5.3.7
6
ABC transporter expression in HMEC-1 cells............................................................... 119
5.4
Summary.............................................................................................................................. 121
5.5
References ........................................................................................................................... 122
Towards an ATPase assay for the human ABCG2 transporter ...................................125
6.1
Introduction ......................................................................................................................... 125
6.2
Materials and Methods ....................................................................................................... 125
6.2.1
Materials...................................................................................................................... 125
6.2.2
Transformation of E. coli ............................................................................................. 126
6.2.3
General procedures for preparation of plasmid DNA ................................................. 127
6.2.3.1
Mini- and Maxi-Prep ................................................................................................ 127
6.2.3.2
Restriction enzyme digestion and dephosphorylation of plasmid ends ................. 127
6.2.3.3
Agarose gel electrophoresis .................................................................................... 127
6.2.3.4
Purification of PCR products and recovery of DNA fragments from agarose gels .. 128
6.2.4
Preparation of the S-ABCG2 construct via sequential overlap extension PCR ............ 128
6.2.4.1
PCR 1a for the S-ABCG2 construct........................................................................... 130
6.2.4.2
PCR 1b for the S-ABCG2 construct .......................................................................... 131
6.2.4.3
PCR 2 for the S-ABCG2 construct............................................................................. 131
6.2.5
Preparation of the S-ABCB1 construct ........................................................................ 132
6.2.6
Subcloning of the S-ABCB1 and the S-ABCG2 construct into pVL1392 vector ............ 134
6.2.7
Sf9 insect cell culture and generation of recombinant baculoviruses ........................ 135
6.2.8
Recombinant transporter expression in pVL1392/S-ABCG2-infected Sf9 cells........... 136
6.2.8.1
Immunological detection of ABCG2 expression ...................................................... 136
6.2.8.2
Membrane preparation ........................................................................................... 136
6.2.9
6.3
ATPase assay for the human BCRP .............................................................................. 137
6.2.9.1
Principle ................................................................................................................... 137
6.2.9.2
ATPase assay protocol ............................................................................................. 138
Results and discussion ......................................................................................................... 142
6.3.1
Results of DNA sequencing.......................................................................................... 142
6.3.2
ABCG2 expression in Sf9 cells ...................................................................................... 144
6.3.2.1
Infection time .......................................................................................................... 144
6.3.2.2
Glycosylation of the ABCG2 transporter ................................................................. 146
6.3.3
6.4
Optimization, validation and application of the ABCG2 ATPase assay ....................... 148
6.3.3.1
Set-up of the ATPase assay for the human BCRP .................................................... 148
6.3.3.2
Mode of ATPase inhibition ...................................................................................... 150
6.3.3.3
Cholesterol-loaded ABCG2 Sf9 membranes ............................................................ 153
6.3.3.4
Effect of CHAPS on basal and drug-stimulated ABCG2-ATPase activity .................. 154
Summary and conclusions ................................................................................................... 158
Contents
6.5
XIII
References ........................................................................................................................... 159
7
Summary ................................................................................................................163
A
Appendix:
Expression
of
ABCG2
at
the
murine
blood-brain
barrier
-
immunohistochemical investigations .............................................................................167
A.1
Introduction ......................................................................................................................... 167
A.2
Materials and methods ....................................................................................................... 168
A.2.1
Drugs and chemicals .................................................................................................... 168
A.2.2
Paraffin embedding and sectioning............................................................................. 168
A.2.3
Immunoperoxidase staining ........................................................................................ 168
A.3
Results ................................................................................................................................. 169
A.4
Summary and conclusions ................................................................................................... 171
A.5
References ........................................................................................................................... 171
XIV
Abbreviations
Abbreviations
AB
ABCB1
ABCC1
ABCG2
ADP
AM
APS
aq.
ATCC
ADP
ATP
AUC
BBB
BCNU
BCRP
BMDP
bp
BSA
CBAVD
cDNA
CHAPS
CHI
CIN
CLSM
CNS
CPD
Da
DAB
DJS
DMEM
DMSO
DNA
Doxo
DTT
EC
ECL
E. coli
EDTA
EGTA
EMEM
ESI
Etp
FACS
antibody
ATP-binding cassette transporter, subfamily B, member 1
ATP-binding cassette transporter, subfamily C, member 1
ATP-binding cassette transporter, subfamily G, member 2
adenosine diphosphate
acetoxymethylester
ammonium peroxodisulfate
aqueous
American Type Culture Collection
adenosine diphosphate
adenosine triphosphate
area under the concentration-time curve
blood-brain barrier
1,3-bis(2-chloroethyl)-1-nitrosourea (carmustine)
Breast Cancer Resistance Protein (= ABCG2 transporter)
Brain Multidrug Resistance Protein
base pair(s)
bovine serum albumin
Congenital bilateral aplasia of vas deferens
copy DNA (desoxyribonucleic acid)
3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate
Chromatographic Hydrophobicity Index
Chromosome Instability
Confocal Laser Scanning Microscopy
central nervous system
citrate-phosphate-dextrose
Dalton
3,3'-diaminobenzidine
Dubin-Johnson Syndrom
Dulbecco’s Modified Eagle Medium
dimethyl sulfoxide
desoxyribonucleic acid
doxorubicin
Dithiothreitol
endothelial cells
enhanced chemiluminescence
Escherichia coli
ethylene diamine tetraacetic acid
ethylene glycol tetraacetic acid
Eagle’s Minimum Essential Medium
Electron-Spray-Ionization
etoposide
Fluorescence Activated Cell Sorter
Abbreviations
FCS
FLAG
FLI
FSC
FTC
g
GBM
GeoMean
H33342
HDL
HE
HMEC
HPLC
HPLC-MS
HRP
HSA
HUGO
IAM
IEP
IHC
Imax
i.p.
IC50
IgG
kb
LB
logD
logP
Luc2
mAU
MDR
MDR1
MeCN
Mito
MG
MOPS
MRP1
MTX
MWCO
MXR
NBD
Pac
PBCEC
PBS
PBS-T
fetal calf serum
octapeptide epitope for the labelling of proteins
fluorescence in vivo imaging
forward scatter
fumitremorgin C
gravitation acceleration
Glioblastoma multiformae
geometric mean value
Hoechst 33342 dye
high density lipoprotein
haematoxylin-eosin staining
human microvascular endothelial cells
high performance (pressure) liquid chromatography
high performance (pressure) liquid chromatography - mass spectrometry
horseradish peroxidase
human serum albumin
Human Genome Nomenclature Committee / Human Genome Organization
Immobilized artifical membrane
Isoelectric point
Immunohistochemistry
maximal inhibitory effect
intraperitoneal
concentration of inhibitor required to give 50% inhibition of activity
immunoglobulin G
kilobase
lysogeny broth (for cultivation of E. coli)
logarithm of the octanol/ water distribution coefficient as a function of pH
logarithm of the octanol/ water partition coefficient
luciferase2 enzyme
milli-absorbance units of peak area
multidrug resistance
multidrug resistance protein 1 (= ABCB1 transporter)
acetonitrile
mitoxantrone
Masson-Goldner staining
3-morpholinopropane-1-sulfonic acid
Multidrug Resistance Associated Protein 1 (= ABCC1 transporter)
methotrexate
Molecular Weight Cut Off
Mitoxantrone resistance protein (= ABCG2 transporter)
nucleotide binding domain
paclitaxel
porcine capillary endothelial cells
phosphate buffered saline
PBS supplemented with 0.1 % triton X-100
XV
XVI
PCR
PE
PEG
PFA
P-gp
pH
PhA
PHHI
Pi
PMSF
PPB
psi
PXE
RAMEB
rcf
RFI
RFU
RP
rpm
rt
S
s.c.
SDS
SDS-PAGE
SEM
Sf9
SOC
SSC
TAE
TBS-T
TEMED
TEER
TFA
TMD
TMEP
Topo
Tris
Trp
Tyr
U
UV
Vbl
WB
WHO
wt
Abbreviations
polymerase chain reaction
phycoerythrin
polyethylene glycol
paraformaldehyde
P-glycoprotein (= ABCB1 transporter)
negative logarithm of the hydrogen ion concentration
pheophorbide a
persistent hyperinsulinemic hypoglycemia of infancy
inorganic phosphate
phenylmethylsulfonyl fluoride
plasma protein binding
pounds per square inch (1 psi = 0.069 bar)
Pseudoxanthoma elasticum
randomly methylated-β-cyclodextrin
relative centrifugal force
relative fluorescence intensity
relative fluorescence unit
reversed phase
revolutions per minute
room temperature
signal peptide sequence
subcutaneous
sodium dodecyl sulfate
SDS polyacrylamide gel electrophoresis
standard error of the mean
insect cell line of Spodoptera frugiperda
salt optimized + carbon broth (for transformation of E. coli)
sideward scatter
tris-acetat-EDTA buffer
Tris buffered saline Tween20 buffer
N,N,N',N'-tetramethylethylenediamine
transendothelial electrical resistance
trifluoroacetic acid
transmembrane domain
Tris-mannitol-EDTA-PMSF buffer
topotecan
tris(hydroxymethyl)aminomethane
tryptophan
tyrosine
enzyme unit
ultraviolet
vinblastine
Western blot
World Health Organization
wildtype
Chapter 1
1
General introduction
1.1
The ABC protein superfamily
The ATP-binding cassette (ABC) protein superfamily is one of the largest transporter gene families in
living organisms. So far, 48 genes encoding ABC proteins have been identified in humans.
The Transporter Classification (TC) Database, a classification system for membrane transport
proteins, approved by the International Union of Biochemistry and Molecular Biology (IUBMB),
divides the ABC family (TC# 3.A.1) into importers and exporters and subdivides the exporters into
three evolutionarily independent distinct superfamilies [Wang et al., 2009]. These have been
designated ABC1, ABC2 and ABC3 as they evolved along three different pathways of origin, arising
from triplication of a primordial 2 transmembrane α-helical segment (TMS), duplication of a threeTMS-encoding genetic element or from a 4 TMS precursor yielding either eight- or ten-TMS proteins,
respectively. Within the ABC superfamily there are dozens of subfamilies corresponding to their
substrate specificities.
According to a standard nomenclature recommended by to the Human Genome Organization
(HUGO), the ABC proteins are divided into seven subfamilies (ABCA-ABCG) based on their sequence
homology and domain organization [Dean et al., 2001], in particular the monophyletic ATP
hydrolyzing
constituents.
A
comprehensive
list
is
available
at
http://www.genenames.org/genefamilies/ABC.
ABCB1 (P-glycoprotein, P-gp; MDR1), ABCC1 (multidrug resistance-associated protein 1, MRP1) and
ABCG2 (Breast Cancer Resistance Protein, BCRP; mitoxantrone resistance protein, MXR) are the most
intensely studied representatives of all ABC transporters and play a major role in the transport of
drugs across biological barriers and the development of multidrug resistance (MDR) in cancer cells
[Borst and Elferink, 2002].
2
1.2
Chapter 1
Physiological and pharmacological functions of ABC
transporters
The ABC efflux pumps use the energy of ATP hydrolysis to transport a huge variety of molecules
against the concentration gradient across biological membranes [Deeley et al., 2006]. In our cellular
defense system, the location of ABC transporters in the gastrointestinal tract represents the first
defense mechanism against numerous ingested xenobiotics. As this also includes numerous orally
administered drugs, ABC proteins, especially P-gp, play a major role in limiting the uptake and, thus,
the bioavailability of drugs [Sarkadi et al., 2006]. ABC transporters are furthermore involved in
limiting the access of various compounds to sensitive compartments like the testes, the placenta or
the brain, thereby protecting these organs from toxic agents. Classically exemplified by ABCB1,
ABCC1 and ABCG2, respectively, it became clear that these plasma membrane proteins are key
players in protecting the organism from xenobiotics and toxines by preventing their uptake and
procuring their elimination.
Mutations of ABC proteins are responsible for the appearance of diseases such as Dubin-Johnson
syndrome (DJS; ABCC2), mucoviscidosis (ABCC7) or pseudoxanthoma elasticum (PXE; ABCC6)
[Chassaing et al., 2005; Deeley et al., 2006; Gadsby et al., 2006; Burke et al., 2008]. Furthermore, the
cholesterol efflux regulatory protein (CERP; ABCA1) mediates the transport of cholesterol and
phospholipids from cells to HDL apolipoproteins, such as apoA-I [Oram et al., 2008]. Mutations of the
ABCA1 protein led to hypoalphalipoproteinemia (Tangier disease), an inherited disorder
characterized by a reduced circulation of HDL and a risk factor for cardiovascular disease
[Negi et al., 2013].
1.2.1
Structures and cellular mechanisms of ABC transporters
In eukaryotes, the ABC full-transporter is supposed to encode four domains consisting of two
cytosolic ATP-binding segments, also known as nucleotide-binding domains (NBDs) and two
transmembrane domains (TMDs). Half-size transporters contain only one NBD and one TMD forming
either homodimers or heterodimers to create a functional transporter and are found in both
prokaryotes and eukaryotes. The TMDs contain 6 hydrophobic segments: membrane-spanning
α-helices which form the TM channel and provide the specificity for the substrate (Figure 1.1).
3
e.c.
ABCG2
i.c.
NBD
C
N
e.c.
ABCB1
i.c.
N
NBD
NBD
C
N
e.c.
ABCC1
i.c.
NBD
NBD
C
Figure 1.1: Topology and domain arrangement of the ABC transporters ABCG2, ABCB1 and ABCC1, respectively. ABCG2 is a
half-size transporter and consists of six membrane-spanning segments and one NBD on the N-terminal side of the TMD.
Both the ABCB1 and ABCC1 transporter comprise two ATP-binding sites. ABCC1 additionally contains a N-terminal domain
composed of five transmembrane helices, the physiological relevance of which is widely unknown [Deeley et al., 2006];
illustration adapted from [Gillet et al., 2007] with alterations.
The ATP-binding domains contain three highly conserved motifs: Walker A and B motifs for
nucleotide binding [Walker et al., 1982], separated by 90-120 amino acids and a signature ‘C’ motif
located upstream of the Walker B site and unique to the ABC superfamily [Dean et al., 2001; Sharom,
2008].
Binding of a substrate by an ABC transporter induces a conformational change in the TMDs which is
then transmitted to the NBDs forming a dimer responsible for ATP binding and hydrolysis as well as
for substrate translocation [Higgins, 2001; Damas et al., 2011] with intracellular ADP and inorganic
phosphate (Pi) as end products (Figure 1.2). Many fundamental questions concerning substrate
4
Chapter 1
binding and the following translocation processes, including a detailed mechanism of ATP hydrolysis
[Jones and George, 2013], are still unanswered. Latest findings, however, suggest that two ATPs are
required for the formation of a NBD dimer [Zoghbi and Altenberg, 2014].
Figure 1.2: Schematic molecular mechanism of substrate binding and ATP hydrolysis by ABC transporters (export pumps);
TMD = transmembrane domain, NBD = nucleotide-binding domain, S = substrate, ATP = adenosine triphosphate,
ADP = adenosine diphosphate, Pi = inorganic phosphate.
1.2.2
ABC transporters and the phenomenon of multidrug
resistance
Cancer chemotherapy is compromised by a phenomenon known as multidrug resistance (MDR),
defined as the cross-resistance or insensitivity of cancer cells to numerous drugs which are often
structurally or functionally unrelated [Ullah, 2008]. MDR is often the ultimate cause of failure of
cytotoxic drugs in cancer chemotherapy. Many mechanisms of drug resistance, such as activation of
detoxifying systems, evasion of the apoptotic control, changes in cellular repair mechanisms or
activation of drug efflux pumps have been identified [Gillet and Gottesman, 2010]. Based on this
knowledge, strategies to overcome drug resistance and to increase the efficacy of cancer
chemotherapy have been developed. Thereby, the inhibition or modulation of cellular efflux pumps
to reverse drug resistance has been subject of research for decades.
Despite the protective detoxifying function of ABC transporters, several complications arise in cancer
therapy as the overexpression of ABC proteins directly leads to multidrug resistance. One of the most
5
important phenomena of MDR in cancer is an increased efflux of cytostatics from cancer cells
mediated by transporters of the ABC protein family.
Stimulated by the identification of the MDR1 gene [Juliano and Ling, 1976], initially many clinical
trials in cancer therapy focused on the ABCB1 transporter. P-gp expression was considered to be
solely responsible for the phenomenon of MDR. However, during the last two decades it became
clear that ABCB1 is not the only member of the ABC transporter family which is involved in the
resistance against clinically relevant chemotherapeutics. After the discovery of MRP1 (ABCC1) and its
affiliation to the ABC transporter family [Cole et al., 1992], additional proteins contributing to
chemoresistance were identified, including the Breast Cancer Resistance Protein.
6
Chapter 1
Table 1.1: List of selected human ABC genes, their physiological function and localization as well as
anticancer drugs accepted as substrates by the corresponding efflux pump.a
HUGO
nomenclature
Synonyms,
symbols
TC
identification
number
Localization
Examples of
cytostatics as
substrates
Function,
involvement in
human disease
ABCB1
MDR1,
P-gp, PGY1
3.A.1.201.1
BBB, kidney,
intestine,
placenta
taxanes,
epipodophyllotoxines,
anthracyclines,
vinca alkaloids
multidrug
resistance
ABCB4
MDR3,
PGY3,
MDR2
3.A.1.201.3
liver
paclitaxel,
vinblastine
progressive
familial
intrahepatic
cholestasis (PFIC)
ABCC1
MRP1,
GS-X
3.A.1.208.8
lung, testes,
kidney
doxorubicin,
etoposide,
vincristine,
methotrexate
(MTX)
multidrug
resistance
ABCC2
MRP2,
cMOAT
3.A.1.208.2
liver,
intestine,
kidney
MTX, etoposide,
doxorubicin,
vincristine,
cisplatin
multidrug
resistance,
Dubin-Johnson
syndrome
ABCC6
MRP6,
MLP1
3.A.1.208.10
kidney, liver
Pseudoxanthoma
elasticum
ABCC7
CFTR,
MRP7
3.A.1.202.1
exocrine
tissues
cystic fibrosis
(mucoviscidosis)
ABCG2
MXR,
BCRP
3.A.1.204.2
placenta,
breast, BBB,
intestine
mitoxantrone,
topotecan,
anthracyclines
multidrug
resistance
a
with modifications according to: [Dean et al., 2001; Borst and Elferink, 2002; Gottesman et al., 2002; Deeley et al., 2006;
Szakacs et al., 2008].
1.2.3
ABC Transporters expressed at the blood-brain barrier
The blood-brain barrier (BBB) is a diffusion barrier formed by endothelial cells between the blood
circulation and the central nervous system (CNS). The BBB controls cerebral homeostasis, protects
the CNS from environmental toxins and restricts the access of pharmacologically active compounds
to the brain. Therefore, various transport and carrier systems are expressed at the BBB among which
7
ABC transporters play a crucial role in pharmacotherapy [Shen and Zhang, 2010; ElAli and Hermann,
2011] as they limit the brain uptake of numerous centrally active compounds [Begley, 2004].
basolateral
Endothelial cells
ABCG2
ABCB1
ABCG2
apical
Blood
ABCB1
ABCG2
ABCB1
Figure 1.3: Schematic illustration of ABCB1 and ABCG2 transporter expression at the luminal membrane of brain capillary
endothelial cells at the blood-brain barrier. ABCB1 and ABCG2 substrates (drugs) are pumped from the basolateral to the
blood side of brain capillaries, resulting in a decreased uptake into the brain. Illustration adapted from [Shukla et al., 2011]
with modifications.
Thereby, the chemotherapy of malignant central nervous system (CNS) tumors is compromised due
to the expression of ABC proteins at the blood-brain barrier namely ABCB1 and ABCG2 [Cooray et al.,
2002; Nies et al., 2004; Miller and Cannon, 2013], restricting the access of many potent cytostatics
such as anthracyclines, epipodophyllotoxines or taxanes [Hermann and Bassetti, 2007] to the brain,
provided that these compounds are substrates of the respective pumps.
The phenomenon of multidrug resistance is a major obstacle in the treatment of malignancies of the
CNS. Strategies for drug delivery to the brain by circumventing these ABC transporters are of great
importance.
By analogy with an approach described for ABCB1 [Fellner et al., 2002; Hubensack et al., 2008],
co-administration of ABCG2 modulators forms an attractive strategy to overcome MDR tumors and
to enhance the penetration of cytostatics into the brain to improve chemotherapy of malignancies in
the CNS [Breedveld et al., 2006].
8
1.3
Chapter 1
References
Begley, D. J. ABC transporters and the blood-brain barrier. Curr. Pharm. Des. 2004, 10(12), 12951312.
Borst, P. and Elferink, R. O. Mammalian ABC transporters in health and disease. Annu. Rev. Biochem.
2002, 71, 537-592.
Breedveld, P., Beijnen, J. H. and Schellens, J. H. Use of P-glycoprotein and BCRP inhibitors to improve
oral bioavailability and CNS penetration of anticancer drugs. Trends Pharmacol. Sci. 2006,
27(1), 17-24.
Burke, M. A., Mutharasan, R. K. and Ardehali, H. The sulfonylurea receptor, an atypical ATP-binding
cassette protein, and its regulation of the KATP channel. Circ. Res. 2008, 102(2), 164-176.
Chassaing, N., Martin, L., Calvas, P., et al. Pseudoxanthoma elasticum: a clinical, pathophysiological
and genetic update including 11 novel ABCC6 mutations. J. Med. Genet. 2005, 42(12), 881892.
Cole, S. P., Bhardwaj, G., Gerlach, J. H., et al. Overexpression of a transporter gene in a multidrugresistant human lung cancer cell line. Science 1992, 258(5088), 1650-1654.
Cooray, H. C., Blackmore, C. G., Maskell, L., et al. Localisation of breast cancer resistance protein in
microvessel endothelium of human brain. Neuroreport 2002, 13(16), 2059-2063.
Damas, J. M., Oliveira, A. S., Baptista, A. M., et al. Structural consequences of ATP hydrolysis on the
ABC transporter NBD dimer: molecular dynamics studies of HlyB. Protein Sci. 2011, 20(7),
1220-1230.
Dean, M., Hamon, Y. and Chimini, G. The human ATP-binding cassette (ABC) transporter superfamily.
J. Lipid Res. 2001, 42(7), 1007-1017.
Deeley, R. G., Westlake, C. and Cole, S. P. Transmembrane transport of endo- and xenobiotics by
mammalian ATP-binding cassette multidrug resistance proteins. Physiol. Rev. 2006, 86(3),
849-899.
ElAli, A. and Hermann, D. M. ATP-binding cassette transporters and their roles in protecting the
brain. Neuroscientist 2011, 17(4), 423-436.
Fellner, S., Bauer, B., Miller, D. S., et al. Transport of paclitaxel (Taxol) across the blood-brain barrier
in vitro and in vivo. J. Clin. Invest. 2002, 110(9), 1309-1318.
Gadsby, D. C., Vergani, P. and Csanady, L. The ABC protein turned chloride channel whose failure
causes cystic fibrosis. Nature 2006, 440(7083), 477-483.
Gillet, J. P., Efferth, T. and Remacle, J. Chemotherapy-induced resistance by ATP-binding cassette
transporter genes. Biochim. Biophys. Acta. 2007, 1775(2), 237-262.
Gillet, J. P. and Gottesman, M. M. Mechanisms of multidrug resistance in cancer. Methods Mol. Biol.
2010, 596, 47-76.
Gottesman, M. M., Fojo, T. and Bates, S. E. Multidrug resistance in cancer: role of ATP-dependent
transporters. Nat. Rev. Cancer 2002, 2(1), 48-58.
9
Hermann, D. M. and Bassetti, C. L. Implications of ATP-binding cassette transporters for brain
pharmacotherapies. Trends Pharmacol. Sci. 2007, 28(3), 128-134.
Higgins, C. F. ABC transporters: physiology, structure and mechanism--an overview. Res. Microbiol.
2001, 152(3-4), 205-210.
Hubensack, M., Müller, C., Höcherl, P., et al. Effect of the ABCB1 modulators elacridar and tariquidar
on the distribution of paclitaxel in nude mice. J. Cancer Res. Clin. Oncol. 2008, 134(5), 597607.
Jones, P. M. and George, A. M. Mechanism of the ABC transporter ATPase domains: catalytic models
and the biochemical and biophysical record. Crit. Rev. Biochem. Mol. Biol. 2013, 48(1), 39-50.
Juliano, R. L. and Ling, V. A surface glycoprotein modulating drug permeability in Chinese hamster
ovary cell mutants. Biochim. Biophys. Acta. 1976, 455(1), 152-162.
Miller, D. S. and Cannon, R. E. Signaling Pathways that Regulate Basal ABC Transporter Activity at the
Blood-Brain Barrier. Curr. Pharm. Des. 2013.
Negi, S. I., Brautbar, A., Virani, S. S., et al. A novel mutation in the ABCA1 gene causing an atypical
phenotype of Tangier disease. J. Clin. Lipidol. 2013, 7(1), 82-87.
Nies, A. T., Jedlitschky, G., Konig, J., et al. Expression and immunolocalization of the multidrug
resistance proteins, MRP1-MRP6 (ABCC1-ABCC6), in human brain. Neuroscience 2004,
129(2), 349-360.
Oram, J. F., Wolfbauer, G., Tang, C., et al. An amphipathic helical region of the N-terminal barrel of
phospholipid transfer protein is critical for ABCA1-dependent cholesterol efflux. J. Biol. Chem.
2008, 283(17), 11541-11549.
Sarkadi, B., Homolya, L., Szakacs, G., et al. Human multidrug resistance ABCB and ABCG transporters:
participation in a chemoimmunity defense system. Physiol. Rev. 2006, 86(4), 1179-1236.
Sharom, F. J. ABC multidrug transporters: structure, function and role in chemoresistance.
Pharmacogenomics 2008, 9(1), 105-127.
Shen, S. and Zhang, W. ABC transporters and drug efflux at the blood-brain barrier. Rev. Neurosci.
2010, 21(1), 29-53.
Shukla, S., Ohnuma, S. and Ambudkar, S. V. Improving cancer chemotherapy with modulators of ABC
drug transporters. Curr. Drug Targets 2011, 12(5), 621-630.
Szakacs, G., Varadi, A., Ozvegy-Laczka, C., et al. The role of ABC transporters in drug absorption,
distribution, metabolism, excretion and toxicity (ADME-Tox). Drug. Discov. Today 2008, 13(910), 379-393.
Ullah, M. F. Cancer multidrug resistance (MDR): a major impediment to effective chemotherapy.
Asian Pac. J. Cancer Prev. 2008, 9(1), 1-6.
Walker, J. E., Saraste, M., Runswick, M. J., et al. Distantly related sequences in the alpha- and betasubunits of ATP synthase, myosin, kinases and other ATP-requiring enzymes and a common
nucleotide binding fold. EMBO J. 1982, 1(8), 945-951.
10
Chapter 1
Wang, B., Dukarevich, M., Sun, E. I., et al. Membrane porters of ATP-binding cassette transport s
ystems are polyphyletic. J. Membr. Biol. 2009, 231(1), 1-10.
Zoghbi, M. E. and Altenberg, G. A. ATP binding to two sites is necessary for dimerization of
nucleotide-binding domains of ABC proteins. Biochem. Biophys. Res. Commun. 2014, 443(1),
97-102.
Chapter 2
2
Scope and Objectives
Previously, a series of new tariquidar analogs was synthesized, which were surprisingly identified as
potent and selective ABCG2 modulators with compound UR-ME22-1 as the most promising
representative [Kühnle et al., 2009]. Aiming at higher water solubility, compounds bearing
triethylenglycol groups were prepared to improve efficacy [Ochoa-Puentes et al., 2011]. Indeed, the
maximal response of this new class of 2nd generation modulators clearly increased, but
unfortunately, by analogy with their precursors, in mouse plasma the compounds were enzymatically
cleaved at the central core structure within 10 min.
In continuation of the work of Matthias Kühnle [Kühnle, 2010] and Peter Höcherl [Höcherl, 2010],
one aim of this thesis was to obtain potent and stable ABCG2 modulators with regard to planned
in vivo studies in orthotopic brain tumor xenograft models in nude mice. For this purpose, the
chemical structure of the already characterized 1st and 2nd generation ABCG2 modulators was
modified in close cooperation with the group of Prof. Dr. Burkhard König (Institute of Organic
Chemistry, University of Regensburg).
To explore, if the new 3rd and 4th generation modulators meet the criteria of future tumor
pharmacological studies, in this thesis bio-analytical and functional investigations had to be
performed:

Firstly, the ABCG2 potency and selectivity against other ABC transporters, i.e. ABCB1 and
ABCC1, had to be tested in vitro in fluorescent based microplate assays using ABC
transporter overexpressing cells. To broaden the spectrum of screening methods, alternative
new assays in the 96-well plate format had to be established.

Biopharmaceutical aspects, especially chemical stability, binding to plasma proteins
(albumin) as well as potential cytotoxicity had to be considered.
12
Chapter 2

To get closer to the in vivo situation, selected ABCG2 modulators were tested in a
blood-brain barrier model based on porcine brain capillary endothelial cells in cooperation
with the group of Prof. Dr. H.-J. Galla (Institute of Biochemistry, University of Münster).

To examine functional interactions of the newly developed modulators with the ABCG2
transporter, a colorimetric ATPase assay using membrane preparations from recombinant
baculovirus infected Sf9 cells was aimed at. This assay type should give information whether
the synthesized compounds act as ABCG2 transporter substrates or as inhibitors.
In view of the treatment of malignant brain tumors, different human brain tumor cell types were
characterized with regard to chemosensitivity against established anti-cancer drugs, their growth
kinetics and their ABC transporter expression as well as their tumorigenicity in subcutaneous tumor
xenograft models.
2.1
References
Höcherl, P. New tariquidar-like ABCB1 modulators in cancer chemotherapy: Preclinical
pharmacokinetic / pharmacodynamic investigations and computational studies. PhD thesis,
University of Regensburg, Germany, 2010.
Kühnle, M. Experimental therapy and detection of glioblastoma: investigation of nanoparticles,
ABCG2 modulators and optical imaging of intracerebral xenografts. PhD thesis, University of
Regensburg, Germany, 2010.
Kühnle, M., Egger, M., Müller, C., et al. Potent and selective inhibitors of breast cancer resistance
protein (ABCG2) derived from the p-glycoprotein (ABCB1) modulator tariquidar. J. Med.
Chem. 2009, 52(4), 1190-1197.
Ochoa-Puentes, C., Höcherl, P., Kühnle, M., et al. Solid phase synthesis of tariquidar-related
modulators of ABC transporters preferring breast cancer resistance protein (ABCG2). Bioorg.
Med. Chem. Lett. 2011, 21(12), 3654-3657.
Chapter 3
3
Modulation of the Breast Cancer
Resistance Protein (ABCG2)
3.1
Introduction
ATP-binding cassette (ABC) transporters prevent the entry of a broad variety of anti-cancer drugs
into cells, which limits their therapeutic value. In cancer, the expression of ABC transporters such as
ABCB1 (P-gp), ABCC1 (MRP1) or ABCG2 (BCRP) is associated with the phenomenon of multidrug
resistance (MDR) [Gottesman et al., 2002; Han and Zhang, 2004]. Hence, chemotherapy of malignant
brain tumors is compromised by efflux pumps expressed at the blood-brain barrier (BBB), especially
by ABCB1 and ABCG2, as numerous cytostatics are substrates of these proteins (cf. Chapter 1).
3.1.1
The ABCG2 transporter
The human ABCG2 protein is a 655 amino acid polypeptide and, like all members of the ABCG
subfamily,
a
half-size
transporter
forming
homodimers
in
the
plasma
membrane
[Sarkadi et al., 2004]. As ABCG2 is highly expressed at the placenta [Staud et al., 2006], the mammary
gland, the intestine and at the blood-brain barrier [Cooray et al., 2002; Robey et al., 2007] it is
thought to play an important role in normal tissues in protecting these organs from potentially toxic
xenobiotics. Although its role in health is not fully clear, the protection of the fetus from toxic organic
anions by excreting metabolites into the maternal circulation seems to be a major task of the ABCG2
transporter [St-Pierre et al., 2000; Fetsch et al., 2006].
Since ABCG2 was first described in drug-resistant cell lines [Chen et al., 1990; Doyle et al., 1998],
various chemically unrelated cytoststatic agents have been demonstrated to be substrates of the
protein, like the anthracenedione mitoxantrone, the camptothecin-derived aromatic topoisomerase I
inhibitors topotecan and irinotecan as well as the anthracyclines daunorubicin and doxorubicin and
many more [Bates et al., 2001; Doyle and Ross, 2003; Robey et al., 2007]. Overexpression of ABCG2
14
Chapter 3
has also been shown to confer moderate resistance to etoposide and methotrexate, but not to
cisplatin, paclitaxel or vinblastine [Jäger, 2009].
3.1.2
BCRP inhibitors
Co-administration of ABCG2 inhibitors and appropriate cytostatics might be useful to enable the
drugs to cross the BBB and to overcome the treatment refractiveness of malignancies in the CNS by
increasing the drug levels [Breedveld et al., 2006].
Compared to other ABC transporters like P-glycoprotein and MRP1, the number of reported ABCG2
inhibitors is still limited [Ahmed-Belkacem et al., 2006], although the list has grown during the last
years: the diketopiperazine fumitremorgin C (FTC; Figure 3.1), isolated from Aspergillus fumigatus,
was reported first [Rabindran et al., 2000], but neurotoxic effects banned the compound from in vivo
studies. An analogue of the natural compound FTC, Ko143 [Allen et al., 2002], is known as one of the
most potent and selective ABCG2 inhibitors. Elacridar (GF120918) [de Bruin et al., 1999], tariquidar
(XR9576) [Kannan et al., 2011] and their common analog WK-X-34 [Jekerle et al., 2006; Jekerle et al.,
2007] are potent dual ABCB1 and ABCG2 modulators.
Different flavonoids [Zhang et al., 2004; Zhang et al., 2005] and chromone derivatives [Valdameri et
al., 2012; Winter et al., 2013] as well as methoxy-chalcones [Valdameri et al., 2012] were described
as potent ABCG2 inhibitors. Previously, the same group identified resveratrol derivatives (methoxystilbenes) as specific inhibitors of ABCG2 [Valdameri et al., 2012].
A compound called YHO-13177 and its water-soluble diethylaminoacetate prodrug were recently
shown to reverse BCRP-mediated drug resistance in vitro and in vivo [Yamazaki et al., 2011].
However, despite the increased trend in identifying ABCG2 inhibitors, little research has been
conducted with a focus on a possible in vivo application.
Nevertheless, the co-adminstration of anticancer agents with BCRP inhibitors seems a promising
concept to treat brain tumors by enhancing the drug delivery into the brain and will, hopefully,
attract further attention in preclinical and clinical investigations.
During the last years, our group, in corporation with the Institute of Organic Chemistry (Prof. Dr. B.
König, University of Regensburg), was able to identify highly potent inhibitors of ABCG2 by a
consequent straight-forward synthesis strategy optimizing the structure-activity relationships of
recently described BCRP modulators [Kühnle et al., 2009]. Our efforts resulted in a series of
compounds among which are the most potent ABCG2 modulators reported so far
[Ochoa-Puentes et al., 2011; Bauer et al., 2013; Ochoa-Puentes et al., 2013].
Modulation of the Breast Cancer Resistance Protein (ABCG2)
3.2
15
Objective
In continuation of the work of Matthias Kühnle [Egger, 2009; Kühnle, 2010] the development and
in vitro characterization of potent and stable ABCG2 modulators was of major interest. With regard
to structural optimization concerning chemical stability under physiological conditions, recently
synthesized substances [Kühnle et al., 2009] were modified and further analogs were prepared. The
compounds were characterized in view of their inhibitory potency against ABCG2, their toxicity and
the extent of reverting ABCG2 mediated drug resistance. Additionally, the selectivity against ABCB1
and ABCC1 was investigated using a new established fluorescence based calcein-AM assay. The most
potent inhibitors were investigated with respect to their chemical stability in mouse plasma by
means of HPLC-MS analysis as an assessment for planed in vivo studies in nude mice.
3.3
Materials and methods
3.3.1
Drugs and chemicals
Hoechst 33342 (Invitrogen, Karlsruhe, Germany) was dissolved in sterile water to produce a 0.8 mM
working solution. A stock solution of mitoxantrone was performed by diluting Novantron® (Wyeth
Pharma, Münster, Germany) in 70% ethanol to a concentration of 2 mM. Pheophorbide a (PhA;
Frontier Scientific, Logan, UT) was dissolved in DMSO to prepare a 200 µM working solution.
Calcein-AM (4 mM in anhydrous DMSO) and pluronic F127 were obtained from Biotium (Hayward,
CA, USA). Topotecan and vinblastine (vinblastine sulfate; both from Sigma, Munich, Germany) were
diluted in 70% ethanol [100 µM] and stored at 4 °C.
Bovine serum albumin (BSA) and crystal violet were purchased from Serva (Heidelberg, Germany).
PBS (phosphate buffered saline) was made of 8.0 g/L NaCl, 1.0 g/L Na2HPO4 · 2 H2O, 0.20 g/L KCl,
0.20 g/L KH2PO4 and 0.15 g/L NaH2PO4 · H2O. The pH-value was adjusted to 7.4 by using 1 M NaOH.
Loading buffer was made of 120 mM NaCl, 5 mM KCl, 2 mM MgCl2 · 6 H2O, 1.5 mM CaCl2 · 2 H2O,
25 mM HEPES (4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid), 10 mM glucose, pH 7.4.
A solution of 4% (m/v) paraformaldeyde (PFA) in PBS was made by stirring 2 g of PFA per 50 mL total
solution while heating on a magnetic stirrer for approximately 60 min.
If not otherwise stated, chemicals (p.a. quality) were obtained from Merck (Darmstadt, Germany).
Purified water (Milli-Q system, Millipore, Eschborn, Germany) was used throughout.
16
3.3.2
Chapter 3
Test compounds
The ABC-transporter modulators were synthesized by Dr. Michael Egger [Egger, 2009], Dr. Cristian
Ochoa-Puentes, Manuel Bause and Qui Sun in the workgroup of Prof. Dr. König (Institute of Organic
Chemistry, University of Regensburg, Germany) and dissolved in DMSO to a concentration of 10 mM
if possible. All stocks were stored at -20 °C.
In the following, the synthesized ABCG2 modulators are termed as 1st, 2nd, 3rd and 4th generation
modulators, respectively, according to their structural characteristics.
Tariquidar was synthesized in our laboratory according to literature [Dodic et al., 1995; Roe et al.,
1999] with slight modifications [Hubensack, 2005].
Fumitremorgin C (FTC; Merck, Darmstadt, Germany) was also dissolved in DMSO and diluted to a
concentration of 1 mM as well as elacridar, a kind gift of GlaxoSmithKline (Munich, Germany). The
FTC-analog Ko143 was a kind gift of Dr. A. H. Schinkel from the Netherlands Cancer Institute
(Amsterdam, NL). Reversan was purchased from Tocris Bioscience (Bristol, UK) and diluted to a 3 mM
working solution in DMSO.
Figure 3.1 gives an overview of the standard ABC transporter modulators. All synthesized ABCG2
modulators are summarized in Tables 3.1-3.7.
17
Figure 3.1: Structures of the ABCG2 modulators fumitremorgin C (a), its analog Ko143 (b), the dual ABCB1/ABCG2 inhibitors
elacridar (c) and tariquidar (d) and the ABCC1 inhibitor reversan (e).
18
Chapter 3
Table 3.1: Structures of the two most potent 1st generation ABCG2 modulators bearing an amide
core [Kühnle et al., 2009].
Table 3.2: Structures of 2nd generation ABCG2 modulators with variations of R1 and R2
[Ochoa-Puentes et al, 2011].
19
Figure 3.2: Replacement of the central benzamide moiety according to a bioisosteric approach by a biphenyl system, an
rd
th
indole, a quinoline and a trizole core, respectively: 3 and 4 generation modulators.
20
Chapter 3
Table 3.3: Structures of 3rd generation ABCG2 modulators: biphenyl series.
Table 3.4: Structures of 3rd generation ABCG2 modulators: indole series.
21
22
Chapter 3
Table 3.5: Structures of 3rd generation ABCG2 modulators: triazole series.
Table 3.6: Structures of 3rd generation ABCG2 modulators: quinoline series.
23
24
Chapter 3
Table 3.7: Structures of 4th generation ABCG2 modulators: triazole series containing a ketone.
3.3.3
New fluorescent ABCG2 modulators
Selected compounds of the triazole series showed distinctive fluorescence, when diluted with
MeCN/0.05% TFA (aq.) at a ratio of 70/30 for HPLC analysis. Fluorescence spectra were recorded
with a LS50 B spectrofluorimeter from Perkin Elmer (Rodgau, Germany) at a scan rate of 240.
3.3.4
Cell culture
MCF-7/Topo cells, an ABCG2 overexpressing subclone of MCF-7 cells (ATCC® HTB-22™, a human
breast adenocarcinoma cell line), were cultured in Eagle’s Minimum Essential Medium (EMEM;
Sigma, Munich, Germany) containing L-glutamine, 2.2 g/L NaHCO3 and 110 mg/L sodium pyruvate
(Serva, Heidelberg, Germany) supplemented with 10% fetal calf serum (FCS; Biochrom, Berlin,
Germany) and selected as described [Hubensack, 2005]. Briefly: treatment of MCF-7 cells with
increasing concentrations of topotecan (final concentration: 550 nM) over approximately 40 days
yielded sufficient quantities of the ABCG2 transporter.
Human Kb-V1 cells, an ABCB1 overexpressing subclone [Kohno et al., 1988; Hubensack, 2005] of
human Kb cells (ATCC® CCL-17™) were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Sigma,
Munich, Germany) supplemented with 3.7 g/L NaHCO3 and 110 mg/L sodium pyruvate (DMEM++) in
addition to 10% FCS and 330 nM vinblastine to maintain sufficient ABCB1 transporter expression.
MDCKII-MRP1 cells: MDCKII cells (Madin-Darby Canine Kidney cells, strain II; a dog epithelial cell line;
ATCC® CRL-2936™), transfected with human ABCC1 [Bakos et al., 1998; Evers et al., 1998], were a
25
kind gift from Prof. Dr. P. Borst from the Netherland Cancer Institute (Amsterdam, NL). The cells were
cultured in DMEM++ supplemented with 10% FCS. Trypsinization was performed at 37 °C (95% air,
5% CO2) for 20-30 min using 0.1% trypsin / 0.04% EDTA (Roche Diagnostics, Mannheim, Germany).
The human glioblastoma cell line U-118 MG (ATCC® HTB-15™) was maintained in DMEM++ containing
10% FCS.
All cells were cultured in a water-saturated atmosphere (95% air, 5% CO2) at 37 °C in 25-cm2, 75-cm2
or 175-cm2 cell culture flasks purchased from Nunc (Wiesbaden, Germany), Greiner (Frickenhausen,
Germany) or Sarstedt (Nümbrecht, Germany) and passaged every 3-7 days after trypsinization using
0.05% trypsin / 0.02% EDTA if not otherwise mentioned. Cell banking was performed according to
the ‘seed stock concept’ [Hay, 1988]. Cells were routinely monitored for mycoplasma contamination
by PCR (Venor® GeM, Minerva Biolabs GmbH, Berlin), and only mycoplasma negative cultures were
used.
3.3.5
Cell based assays for the determination of ABC transporter
modulation
All assays were performed in the microtiter plate format using 96-well plates (Greiner,
Frickenhausen, Germany) and a GENios Pro microplate reader (TECAN Deutschland GmbH,
Crailsheim, Germany) for fluorescence detection.
3.3.5.1 ABCG2 modulation: Hoechst 33342 and pheophorbide a microplate
assays
The standard protocol was as follows (according to [Kühnle, 2010] with slight modifications):
MCF-7/Topo cells were seeded into 96-well plates at a density of 20,000-25,000 cells/well (total
volume: 100 µL) and incubated over night (95% air, 5% CO2) at 37 °C. The next day, the incubation
medium was removed from the microplate. Afterwards, cells were incubated for 2 hours (37 °C,
5% CO2) with loading suspension: EMEM, supplemented with 10% FCS, L-glutamine, 2.2 g/L NaHCO3,
110 mg/L sodium pyruvate and 8 µM Hoechst 33342 (H33342) or 1 µM pheophorbide a (PhA) in
combination with the test compounds at increasing concentrations (10 nM-100 µM). The structures
of H33342 and PhA are shown in Figure 3.3.
Fumitremorgin C served as positive control at a final concentration of 10 μM, corresponding to 100%
inhibition of H33342 and PhA efflux, respectively. The supernatants were discarded, and the cells
were fixed for 20 min under light protection using 100 µL per well of a 4% paraformaldehyde
solution. Finally, MCF-7/Topo cells were washed twice with 250 µL of PBS per well in order to get rid
26
Chapter 3
of residual dye. Afterwards, cells were overlaid with 100 µL of PBS, and fluorescence was determined
with a GENios Pro microplate reader.
Instrument settings were as follows: measurement mode: fluorescence top; excitation filter
(Hoechst 33342): 340/35, excitation filter (pheophorbide a): 380/10; emission filter (Hoechst 33342):
485/20, emission filter (pheophorbide a): 670/25; number of reads: 10; integration time: 40 μs; lag
time: 0 μs; mirror selection: automatic; plate definition file GRE96ft.pdf; multiple reads per well
(Circle): 3x3; time between move and flash: 50 ms.
On each plate, the optimal gain was calculated by determination of the fluorescence intensity in the
presence of the reference compound fumitremorgin C and the microtiter plates were stored at 4 °C
until cell quantification. To consider unspecific toxic effects of the test compounds during the
incubation phase, the obtained fluorescence values were normalized to the biomass in each well. For
this purpose, the cells were stained with a 0.02% aqueous crystal violet solution (100 μL/well) for
20 min. Excess dye was removed by rinsing the microplates with water for 20 min. Crystal violet
bound by the cells was extracted with 70% ethanol (180 μL/well) while shaking the microplates for
2-4 h. Subsequently, the absorbance as a parameter proportional to cell mass was measured at
590 nm with a TECAN plate reader. For normalization of the fluorescence intensities to the cell mass,
fluorescence values were divided by the corresponding absorbances. All values were corrected with
respect to the cellular uptake of the dye by diffusion (DMSO control value), and the data were
referred to the maximal signal measured in the presence of the reference compound FTC at a
concentration of 10 μM (100% inhibition).
Addition of the modulators at increasing concentrations led to sigmoidal concentration response
curves in semilogarithmic plots. IC50 values were calculated using SigmaPlot 11.0 software (Systat
Software GmbH, Erkrath, Germany) by fitting ‘Four parameter logistic curve’. Errors were expressed
as standard error of the mean (SEM).
Figure 3.3: Chemical structures of the fluorescent BCRP substrates H33342 (a) and pheophorbide a (b).
27
3.3.5.2 Mitoxantrone microplate assay for the determination of ABCG2
modulation
Because of unfavourable fluorescent properties of some synthesized compounds and discrepencies
between H33342 and PhA assay, another assay was established for the determination of BCRP
modulation. With excitation and emission wavelengths far from those of the H33342 dye and the
fluorescent modulators, mitoxantrone (Figure 3.4) was chosen as a fluorescent ABCG2 substrate,
since a flow cytometric mitoxantrone assay protocol already existed [Kühnle, 2010].
In terms of comparability with the H33342 and the PhA assay the mitoxantrone assay was also
established in the 96-well format.
Figure 3.4: Structure of the antineoplastic agent mitoxantrone, which is a fluorescent substrate of ABCG2.
MCF-7/Topo cells were seeded into 96-well plates (density: 20,000 cells/well) and incubated over
night (95% air, 5% CO2) at 37 °C. The next day, the incubation medium was removed, and the cells
were overlaid with loading suspension (EMEM, supplemented with 10% FCS, L-glutamine,
2.2 g/L NaHCO3, 110 mg/L sodium pyruvate and 20 µM mitoxantrone in combination with the test
compound at increasing concentrations) for 3 hours (37 °C, 5% CO2).
Fumitremorgin C served as positive control at a final concentration of 10 μM corresponding to 100%
ABCG2 inhibition. The supernatants were drained, and the cells were immediately washed twice with
250 µL of PBS (without fixation). Afterwards, cells were overlaid with 100 µL of PBS, and the relative
fluorescence intensities were determined with the GENios Pro microplate reader. Time between
washing and the measurement was kept as short as possible to avoid ABCG2-independent efflux of
mitoxantrone.
Instrument settings were as follows: measurement mode: fluorescence top; excitation filter: 612/10;
emission filter: 670/25; number of reads: 10; integration time: 40 μs; lag time: 0 μs; mirror selection:
automatic; plate definition file: GRE96ft.pdf; multiple reads per well (Circle): 3x3; time between
move and flash: 100 ms.
The following cell quantification procedure and the calculation of IC50 values were performed by
analogy with the H33342 assay.
28
Chapter 3
3.3.5.3 Calcein-AM efflux assay for the determination of ABCB1 modulation
The assay was performed as described elsewhere [Höcherl, 2010].
3.3.5.4 Calcein-AM efflux assay for the determination of ABCC1 modulation
MDCKII-MRP1 cells were seeded into flat-bottomed 96-well plates at a density of about
20,000 cells/well. On the following day, cells were washed with loading buffer in order to remove
unspecific serum esterases. Afterwards, cells were incubated with loading suspension (loading
buffer, 5 mg/mL BSA, 1.25 μL/mL pluronic F127 (20% (m/v) in DMSO)) containing 0.5 μM calcein-AM
(Figure 3.5) and the test compound at increasing concentrations (10 nM-100 µM) for 60 min
(37 °C / 5% CO2). Reversan served as positive control at a final concentration of 30 μM corresponding
to 100% ABCC1 inhibition. In general, test compounds were investigated as triplicates, controls as
sextuplicates, respectively.
Subsequently, the loading suspension was discarded, and the cells were fixed with 4% PFA solution
for 20 min. After three washing circles (loading buffer), fixed cells were overlaid with loading buffer,
and relative fluorescence intensities were determined at 535/25 nm at the GENios Pro microplate
reader after excitation at 485/20 nm. The following cell quantification procedure was performed by
analogy with the H33342 assay. The obtained mean fluorescence intensities were related to the
controls and plotted against the various concentrations of test compounds.
TECAN instrument settings: Measurement mode: fluorescence top; number of reads: 10; integration
time: 40 μs; lag time: 0 μs; mirror selection: Dichroic 3 (e.g. Fl.); plate definition file: GRE96ft.pdf;
multiple reads per well (Circle): 3x3; time between move and flash: 100 ms.
Figure 3.5: Structure of nonfluorescent calcein-AM (a), a substrate of ABCB1 and ABCC1, which is converted to
green-fluorescent calcein (b) in live cells after acetoxymethyl ester hydrolysis by intracellular esterases.
3.3.6
29
Chemosensitivity assays
The assays were performed as described previously [Bernhardt et al., 1992] with slight modifications.
In brief: 100 μL/well of a tumor cell suspension, yielding a density of 500 cells per well, were seeded
into 96-well plates and incubated over night at 37 °C / 5% CO2 in a water-saturated atmosphere.
The next day, fresh medium (100 µL/well) was added containing the 500-fold concentrated test
compounds (16 wells per drug concentration) giving a final drug dilution of 1:1,000. On every plate,
untreated cells (solvent DMSO in a dilution of 1:1,000) served as growth control (vehicle), cells
treated with a reference cytostatic (e.g. vinblastine) as a positive control. Incubation of the cells was
stopped after different periods of time by removal of medium and fixation with 2% (v/v)
glutardialdehyde (in PBS). All plates were stored at 4 °C until the end of the experiment and stained
simultaneously with 100 μL/well of a 0.02% (m/v) aqueous crystal violet solution for 20 min.
Subsequently, the trays were rinsed with water for 20 min in order to remove residual dye. Cellassociated crystal violet was redissolved in 70% ethanol (180 μL/well) under shaking for 2-4 hours.
Absorbance was measured at 580 nm using a GENios Pro microplate reader.
Drug effects were expressed as corrected T/C-values for each group according to equation 1,
(e ua on 1)
where T is the mean absorbance of the treated cells, C the mean absorbance of the controls and C0
the mean absorbance of the cells at the time (t=0) when the drug was added.
When the absorbance of treated cells T was less than that of the culture at t=0 (C0), the extent of cell
death was calculated according to the following formula:
(e ua on 2)
SigmaPlot analysis software 11.0 was used for the creation of growth curves.
3.3.7
Chemical stability of ABCG2 modulators in human and
mouse plasma
With respect to future in vivo studies, selected candidates of potent ABCG2 modulators were
investigated for their stability in human and mouse plasma.
For mouse plasma, blood of NMRI (nu/nu) mice was collected by cardiac puncture in deep anesthesia
using heparin-coated syringes. Samples were immediately centrifuged at 4,500 g (Eppendorf
30
Chapter 3
centrifuge 5415R, Eppendorf, Hamburg, Germany) for 7 min, and the supernatant was carefully
removed. After pooling, the plasma was fractioned into small aliquots and stored at -80 °C. Human
CPD (citrate-phosphate-dextrose) plasma was purchased from the Bavarian Red Cross (BRK;
Regensburg, Germany).
3.3.7.1 Determination of the activity of unspecific esterases
The enzymatic activity of unspecific esterases in human and murine plasma samples was determined
spectrophotometrically. Therfore, the development of the colored product o-nitrophenol, formed by
enzymatic cleavage of the chromogenic substrate nitrophenyl butyrate, was determined as a
function of time: 40 µL ortho-nitrophenylbutyrate (Fluka, Munich, Germany) were mixed with 1 mL
of 0.1 M Na2HPO4 buffer (pH 7.2). After warming to 37 °C, 10 µL of murine or human plasma were
added and UV absorbance was measured for 2 min at 414 nm using a Cary 100 UV-Vis
spectrophotometer (Varian, Darmstadt, Germany). Absorbance was plotted against time, and the
linear slope at the beginning of the reaction was determined and used for the calculation of the
volume activity according to the following equation:
(e ua on 3)
VA
= volume activity [U/mL]
A
= absorbance of o-nitrophenol at 414 nm
t
= time [min]
V
= total volume [mL]
ε
= molar absorption coefficient (o-nitrophenol) [L· mol-1 · cm-1]
d
= path length [cm]
v
= plasma volume [mL]
3.3.7.2 Assay procedure
The test compounds were dissolved in DMSO at a concentration of 3 mM. A 1:50 dilution of the
substances with mouse plasma was prepared in 1.5-mL polypropylene reaction vessels (Eppendorf,
Hamburg, Germany). The samples were shortly vortexed and immediately incubated at 37 °C. After
increasing periods of time, aliquots were taken, after the reaction had been stopped by addition of
two volumes of ice-cold acetonitrile (MeCN). For quantitative precipitation of the denatured
proteins, the samples were efficiently vortexed and stored at 4 °C for 30 min. Finally, samples were
31
centrifuged at 14,000 g for 5 min, and the supernatants were transferred to new Eppendorf reaction
vials. For HPLC analysis, the samples were diluted (1:2) with MeCN and stored at -80 °C until
measurement.
3.3.7.3 HPLC and HPLC-MS analysis
3.3.7.3.1 HPLC analysis
Subsequent RP-HPLC analysis was performed with a Waters (Eschborn, Germany) system composed
of a 600S controller and pump, a Waters degasser, a temperature control module, a 717 plus
autosampler and a 2487 UV detector. A Luna RP-18 (Phenomenex, Aschaffenburg, Germany)
analytical column (3 μm, 150 mm x 4.6 mm) thermostatted at 30 °C was used with a flow rate of
1.0 mL/min.
Samples were thawn at room temperature, and 100 μL were injected. UV detection was performed
at 210 nm. During recovery analysis, the applied gradient was as following:
MeCN/0.05% TFA (aq.): 0 min: 15/85, 25 min: 80/20, 26 min: 95/5, 36 min: 95/5, 37 min: 15/85,
45 min: 15/85.
3.3.7.3.2 HPLC-MS analysis
Determination of cleavage products was performed by HPLC-MS analysis using an Agilent 1100
(Agilent Technologies, Palo Alto, CA) HPLC system equipped with a Luna C18, 3 μm, 100 mm x 2 mm
column (Phenomenex, Aschaffenburg, Germany) at 40 °C. Gradient elution with water containing
formic acid (0.1%) and MeCN (0 min: 3%, 15 min: 95%, 17 min: 95%, 17.5 min: 3%, 19 min: 3%) was
performed at a flow rate of 0.4 mL/min with UV detection at 220 nm. The HPLC was coupled to a
Finnigan ThermoQuest TSQ (Triple-Stage-Quadrupol) 7000 ESI mass spectrometer.
3.3.8
ABC transporter modulation in a blood-brain barrier model
To get closer to in vivo conditions at the BBB, selected compound were tested in an in vitro
blood-brain barrier model (Figure 3.6) imitating the physiological situation using endothelial cells,
expressing the brain multidrug resistance protein (BMDP), the porcine homolog to the human ABCG2
gene [Hoheisel et al., 1998; Franke et al., 1999; Franke et al., 2000]. Thereby, the transport of the
ABCG2 substrate 3H-daunorubicin across a ‘membrane’ (cell monolayer of primary cultured brain
capillary endothelial cells) was measured in the presence of selected ABCG2 modulators. The
experiments were performed by Sabine Hüwel at the Institute of Biochemistry (Prof. Dr. H.-J. Galla)
at the University of Münster.
32
Chapter 3
astrocytes
pericyte
microvessel lumen
apical compartment
BMDP
brain
basolateral compartment
endothelium
Figure 3.6: Simplified illustration comparing the in vivo situation at the BBB with the in vitro BBB model. The right panel
shows a schematic cross-section through a brain capillary, the left panel shows the corresponding compartment in the
experimental set-up of the transcellular transport model with a monolayer of porcine brain capillary endothelial cells
expressing the brain multidrug resistance protein (BMDP). Illustration adapted from [Franke et al., 2000].
The transport assay protocol was as following [Eisenblätter et al., 2003]: the isolation of porcine
capillary endothelial cells (PBCEC) has been described elsewhere [Franke et al., 2000]. DMEM/Ham’s
F12 medium (serum-free) containing L-glutamine (0.7 mM), penicillin/streptomycin (100 µg/mL),
gentamicin (100 µg/mL) and hydrocortisone (550 nM) was used throughout with cells cultured on
Transwell™ microporous polycarbonate filter membranes (0.4 µm pore size, 1 cm2 growth area;
Costar, Cambridge, MA) to prepare confluent cell monolayers. By supplementing the medium with
hydrocortisone at physiological concentrations, the endothelial monolayer shows a transendothelial
electrical resistance (TEER) of up to 1800 Ω/cm2 similar to the situation in vivo and low paracellular
permeabilities for sucrose [Levin, 1980; Crone and Olesen, 1982]. The barrier properties were
quantified in terms of TEER determined by AC (alternating current) impedance analysis
[Wegener et al., 2000] . The cells were incubated with the inhibitor at a specific concentration 60 min
prior to the experiment, which was started by replacing the medium on both sides of the cell layer
with fresh medium, containing 0.2 µM 3H-daunorubicin (NEN Life Science Products; 5 Ci/mmol;
185 GBq/mmol) and the inhibitor at the corresponding concentration. After 0.5, 6, 12, 24, 36, 48 and
72 h, samples were taken from each compartment. The radioactivity at 0.5 h was set to 100% for
each compartment. The tightness of the monolayer was monitored by the paracellular flux of
14
C-sucrose (Amersham Biosciences; 740 mCi/mmol; 27 GBq/mmol) from the apical to the
basolateral compartment (2 µCi/mL or 1 µCi per filter). Three independent experiments were
performed with three filters each.
3.4
33
Results and discussion
The following results were published in part [Bauer et al., 2013; Ochoa-Puentes et al., 2013]. Selected
data, statements and figures are quoted with the permission of the corresponding journal.
3.4.1
Inhibition of ABCB1, ABCC1 and ABCG2
General procedure of ABC transporter inhibition assays:
In ABC transporter overexpressing cell lines, a fluorescent substrate of the corresponding ABC
protein is not accumulated but extruded by the efflux pump. In combination with ABC
modulators/inhibitors, the intracellular levels of the dyes increase resulting in higher fluorescence
intensities, which can be directly referred to the extent of ABC transporter inhibition.
Intracellular fluorescence was monitored with a plate reader in a concentration-dependent manner.
The relative fluorescent intensities allowed the construction of sigmoidal concentration response
curves and the calculation of IC50 (respective concentration of the inhibitor required for half-maximal
inhibition) and Imax values (inhibition by the highest concentration of the compound expressed in
percent relative to the control compound).
The synthesized new compounds as well as the reference substances tariquidar, elacridar and Ko143
were investigated for ABCB1, ABCC1 and ABCG2 modulation in a concentration range from 10 nM to
100 µM. Precipitate formation in the loading suspension became visible for a number of modulators
at final concentrations above 10 µM. These concentrations were not included in the calculation of
IC50 and Imax values. Constructed concentration-response curves of selected ABCG2 modulators and
the reference compounds are displayed in Figures 3.7-3.8.
34
Chapter 3
120
ABCG2 inhibition (%)
100
80
Figure 3.7: Inhibition of ABCG2
by the reference compounds
tariquidar
(filled
circles),
elacridar (open circles), Ko143
(open triangles) and compounds
2 (filled triangles) and 4 (filled
squares).
60
40
20
0
0.001
0.01
0.1
1
concentration [µM]
10
100
120
100
80
Figure 3.8: Inhibition of ABCG2
rd
th
by 3
and 4
generation
modulators: 12 (open diamonds),
33 (filled inverted triangles), 42
(filled diamonds), and 52 (open
inverted
triangles),
each
representing the most potent
compound of its series.
60
40
20
0
0.001
0.01
0.1
1
10
100
concentration [µM]
All results of the test compounds and the reference compounds regarding ABC transporter
modulation are summarized in Tables 3.8-3.13. If not otherwise stated, mean values were calculated
from two to three independent experiments performed in sextuplicate (ABCG2) or triplicate (ABCB1
and ABCC1). Maximal inhibitory effects are expressed as percental inhibition caused by the highest
concentration of the compound tested (70 or 100 µM, respectively) relative to the control compound
(100%). Compounds without an effect up to a concentration of 100 µM are designated as ‘inactive’.
35
Table 3.8: Inhibition of ABCG2 and ABCB1 by compounds 1-5 and reference compounds.
ABCG2
Compound
ABCB1
IC50 [nM]a
Imax (%)
IC50 [nM]b
Fumitremorgin C
731 ± 92
100
n.d.
Ko143c
117 ± 53
103 ± 7
inactive
Elacridarc
127 ± 41
63 ± 7
193 ± 18d
Tariquidarc
526 ± 85
69 ± 5
223 ± 8d
1 (UR-ME1-4)c
104 ± 16
50 ± 1
9,450 ± 417d
2 (UR-ME22-1)c
65 ± 8
63 ± 2
>29,000d
3 (UR-COP77)e
183 ± 32
83 ± 5
>50,000
e
130 ± 29
88 ± 3
>50,000
508 ± 191
61 ± 4
>50,000
4 (UR-COP78)
5 (UR-COP134)
e
a
b
Mean values ± SEM calculated from two to three independent experiments performed in triplicate; Calcein-AM
c
d
micorplate assay (unless otherwise indicated); n.d.: not determined; [Kühnle, 2010]; data from flow cytometric
e
calcein-AM assay; [Ochoa-Puentes et al., 2011].
Table 3.9: Inhibition of ABCG2, ABCB1 and ABCC1 by 3rd generation modulators: biphenyl series,
compounds 6-17.
Compound
ABCG2
ABCB1
ABCC1
IC50 [nM]
Imax (%)
IC50 [nM]
Imax (%)
IC50 [nM]
6 (UR-COP142)
1,540 ± 110
92 ± 6
>10,000
119 ± 2
n.d.
7 (UR-COP145)
943 ± 79
87 ± 5
inactive
8 (UR-COP147)
641 ± 175
94 ± 2
4,490 ± 160
9 (UR-COP153)
1,461 ± 169
85 ± 7
inactive
n.d.
10 (UR-COP154)
1,031 ± 170
114 ± 20
inactive
n.d.
11 (UR-COP157)
237 ± 65
67 ± 1
>18,000
105 ± 4
n.d.
12 (UR-COP228)
591 ± 87
109 ± 8
1230 ± 30
21 ± 8
n.d.
13 (UR-COP245)
3,214 ± 1,050
126 ± 15
inactive
n.d.
14 (UR-COP246)
1,100 ± 156
90 ± 2
inactive
n.d.
15 (UR-COP248)
760 ± 67
98 ± 6
5,420 ± 230
52 ± 5
n.d.
16 (UR-COP258)
544 ± 53
112 ± 19
5,990 ± 520
55 ± 6
n.d.
17 (UR-COP270)
839 ± 105
103 ± 3
4,430 ± 300
53 ± 7
inactive
n.d.
100 ± 2
n.d.
36
Chapter 3
Table 3.10: Inhibition of ABCG2, ABCB1 and ABCC1 by 3rd generation modulators: indole series,
compounds 18-33.
Compound
ABCG2
ABCB1
ABCC1
IC50 [nM]
Imax (%)
IC50 [nM]
Imax (%)
IC50 [nM]
18 (UR-COP224)
603 ± 204
56 ± 10
inactive
inactive
19 (UR-COP235)
1,264 ± 345
46 ± 2
inactive
inactive
20 (UR-COP236)
1,185 ± 272
50 ± 7
2,048 ± 47
17 ± 2
inactive
21 (UR-COP237)
683 ± 253
54 ± 4
1,076 ± 106
30 ± 7
inactive
22 (UR-COP238)
2,238 ± 202
22 ± 0
inactive
23 (UR-COP239)
639 ± 20
39 ± 2
1,068 ± 239
13 ± 4
inactive
24 (UR-COP256)
920 ± 145
49 ± 1
1,323 ± 368
28 ± 4
inactive
25 (UR-COP271)a
n.d.
26 (UR-COP240)
384 ± 47
27 (UR-COP251)
inactive
n.d.
n.d.
98 ± 2
inactive
inactive
140 ± 40
99 ± 6
1,341 ± 399
13 ± 4
inactive
28 (UR-COP254)
368 ± 63
97 ± 4
7,906 ± 595
15 ± 0
inactive
29 (UR-COP260)
132 ± 6
121 ± 19
3,940 ± 629
56 ± 10
inactive
30 (UR-COP264)
316 ± 83
96 ± 9
inactive
inactive
31 (UR-COP268)
160 ± 4
99 ± 1
inactive
inactive
32 (UR-COP269)
59 ±11
101 ± 5
7,299 ± 910
14 ± 2
inactive
33 (UR-COP272)
46 ± 1
111 ± 9
3,570 ± 640
36 ± 4
inactive
a
Insufficient stability excluded compound 25 from testing.
Table 3.11: Inhibition of ABCG2, ABCB1 and ABCC1 by 3rd generation modulators: triazole series,
compounds 34-40.
Compound
ABCG2
ABCB1
ABCC1
IC50 [nM]
Imax (%)
IC50 [nM]
Imax (%)
IC50 [nM]
34 (UR-MB19-5)
392 ± 40
92 ± 1
inactive
inactive
35 (UR-MB81-2)
426 ± 38
103 ± 1
inactive
inactive
36 (UR-MB83-2)
168 ± 81
96 ± 1
5,371 ± 1624
68 ± 1
inactive
37 (UR-MB84-3)
137 ± 4
88 ± 10
>35,000
34 ± 8
inactive
38 (UR-MB86-2)
371 ± 15
90 ± 2
>15,000
39 ± 4
inactive
39 (UR-MB95-2)
108 ± 24
97 ± 7
1,270 ± 313
91 ± 11
inactive
40 (UR-MB113-3)
145 ± 35
85 ± 6
inactive
inactive
37
Table 3.12: Inhibition of ABCG2, ABCB1 and ABCC1 by 3rd generation modulators: quinoline series,
compounds 41-49.
Compound
ABCG2
ABCB1
ABCC1
IC50 [nM]
Imax (%)
IC50 [nM]
Imax (%)
IC50 [nM]
41 (UR-QQS-1)
536 ± 160
77 ± 1
2,709 ± 158
41 ± 15
n.d.
42 (UR-QQS-2)
1,043 ± 53
107 ± 7
inactive
n.d.
43 (UR-QQS-3)
602 ± 44
60 ± 0
inactive
n.d.
44 (UR-QQS-4)
851 ± 93
81 ± 14
inactive
n.d.
45 (UR-QQS-5)
1,467 ± 568
70 ± 5
inactive
n.d.
46 (UR-QQS-6)
1,167 ± 214
61 ± 7
>15,000
47 (UR-QQS-7)
1,820 ± 861
75 ± 7
inactive
48 (UR-QQS-10)
904 ± 45
58 ± 3
8,571 ± 211
49 (UR-QQS-9)
513 ± 101
53 ± 6
inactive
25 ± 2
n.d.
n.d.
21 ± 1
n.d.
n.d.
Table 3.13: Inhibition of ABCG2, ABCB1 and ABCC1 by 4th generation modulators: triazole series
bearing a ketone, compounds 50-52.
Compound
ABCG2
ABCB1
ABCC1
IC50 [nM]
Imax (%)
IC50 [nM]
Imax (%)
50 (UR-MB107-6)
48 ± 10
98 ± 11
inactive
inactive
51 (UR-MB108-4)
64 ± 3
95 ± 3
inactive
inactive
52 (UR-MB136-4)
55 ± 14
110 ± 17
2,551 ± 57
156 ± 2
IC50 [nM]
inactive
Elacridar strongly inhibited the ABCG2 and the ABCB1 transporter, although without preference for
one of the two targets, whereas tariquidar was about four times less potent at ABCG2, but almost
equipotent with elacridar at ABCB1. The ABCG2 inhibitor Ko143, inactive at ABCB1, showed an IC50
value of 117 nM with a maximal inhibitory effect comparable to that of fumitremorgin C. Compounds
1 and 2 were comparable to Ko143 regarding potency and selectivity, but produced a lower maximal
response. Aiming at structural optimization of potent and selective ABCG2 inhibitors like 2,
compounds bearing triethyleneglycol ether groups at the tetrahydroisoquinoline moiety were
synthesized (compounds 3-5). Compounds 3 and 4 were comparable to UR-ME22-1 in potency, but
considerably more efficient (maximal inhibition of 83% and 88% vs. 63%, relative to FTC). These
results support the hypothesis that poor solubility of the ABCG2 modulators 1 and 2 in aqueous
media is the factor limiting efficacy.
Compounds such as UR-COP78 (4) are highly potent and selective ABCG2 modulators, but prone to
rapid enzymatic cleavage (cf. section 3.4.5.2) at the central benzanilide moiety. In search for more
stable analogs, according to a bioisosteric approach, new series of potential modulators, bearing a
38
Chapter 3
biphenyl, a quinoline, an indole and a triazole core, respectively, were prepared by conventional or
solid phase synthesis.
All newly synthesized compounds showed a marked preference for ABCG2 inhibition.
Within the biphenyl series compounds 7, 9, 10, 13 and 14 were completely inactive at the ABCB1
transporter, whereas the IC50 values of compounds 12 and 16 (~600 nM) were comparable to that of
tariquidar, the maximal inhibitory effects were significantly higher, achieving 109% and 112%,
respectively, relative to FTC. Except for 11, all modulators show a maximal ABCG2 inhibition of 90%
or higher. Compound 12, the most potent - but not completely selective - BCRP modulator in this
series, showed a maximum inhibition at the ABCB1 transporter of only 20%.
The potency of compounds 13-16, bearing an ethylene linker, was not increased compared to that of
the lower homolog 12, suggesting that the distance between the tetrahydroisoquinoline core and the
biphenyl motif may be varied within this class of ABCG2 modulators. Compounds 12 and 16, indeed,
showed higher IC50 values compared to all reference compounds, but, with respect to the maximal
inhibitory effects, they were superior to tariquidar, elacridar, compounds 1-5 and comparable to
Ko143.
Subsequent compounds of the indole, quinoline and triazole series, were designed with a
quinoline-2-carboxamido substituent at the central benzamide.
Compounds 18-33 represent the indole series. Initial problems with deprotection during synthesis led
to compounds 18-24, the N-mesylated precursors of 25-33. In general, the N-mesylated indoles
showed higher IC50 values and lower maximal responses (~50%) compared to the unprotected
analogs, suggesting that the unsubsituted NH-group of the indole core is essential for ABCG2
inhibition. Most of the indoles 25-33 inhibited the ABCG2 transporter at low nanomolar
concentrations with maximal inhibitory effects of 100% or even surpassing that of FTC. Compounds
26, 30 and 31 were selective for ABCG2, whereas the other compounds showed also activity at the
P-glycoprotein with IC50 values in the micromolar range (e.g. selectivity ratio of compound 32: 124 in
favor of ABCG2). Compounds 29, 32 and 33 were the most potent BCRP modulators of this series
with IC50 values of 132 nM, 59 nM 46 nM, respectively, and maximal inhibitory effects higher than
100%. These indoles were superior to tariquidar, FTC and even to Ko143, the most potent ABCG2
transporter inhibitor reported so far. None of the compounds showed activity at the ABCC1
transporter. In general, the potencies of compounds with a methylene linker (n=1) at the indole core
were comparable to those of compounds with an ethylene bridge (n=2).
Compounds 34-40 and 50-52 (triazole series) produced similar results with IC50 values in the low
nanomolar range and potencies comparable to fumitremorgin C (100%). Together with 32 and 33,
compounds 50-52 are the most potent ABCG2 modulators reported so far.
39
Compounds 39 and 52 also showed distinct activity at the P-gp and work as potent dual
ABCB1/ABCG2 transporter modulators with a maximal ABCB1 inhibition of 91% and 156% compared
to tariquidar [1 µM], respectively.
The introduction of one or two triethylene glycol chains at the tetrahydroisoquinoline core in the
series of 3rd generation modulators, in general, on the one hand increased the inhibitory activity
compared to UR-ME22-1 (2), probably due to improved water solubility, on the other hand ABCG2
selectivity was decreased.
In another synthesis approach, a quinoline instead of an indole was introduced as a putative anilide
bioisostere resulting in compounds 41-49. Compounds 41 and 42, bearing one or two triethylenglycol
chains, respectively, were indeed the two most potent representatives of this series, but in general,
the quinoline bearing compounds turned out to be hardly soluble and less potent compared to the
indole series regarding ABCG2 modulation, suggesting that the extension of the indole ring system to
a quinoline reduces ABCG2 inhibition.
Compounds 44-49 contain linkers with various basic heterocycles instead of a methoxy group or a
triehylenglycol chain and were generally less potent and efficient in comparison to compounds of the
indole and the triazole series.
3.4.2
Fluorescence properties of selected ABCG2 modulators
Besides modulators 38, 51 and 52 only compounds 35 and 36 showed weak fluorescence (spectra
not shown) in mobile phase with an emission maximum at 470 nm after excitation at 370 nm
(Figures 3.9-3.10).
40
Chapter 3
500
120
A
B
100
relative intensity
relative intensity
400
300
200
100
80
60
40
20
0
0
300
350
400
450
400
wavenlength [nm]
450
500
wavelength [nm]
550
600
Figure 3.9: A) Excitation spectrum (λem: 480 nm; Ex Slit: 5; Em Slit: 10) of compound 51 [1 µM] in MeCN/0.05% TFA (aq.) at a
ratio of 70/30; B) Emission spectrum (Ex Slit: 10, Em Slit: 5) of compound 51 [1 µM] after excitation at 310 nm (blue) and
370 nm (red), respectively.
120
120
B
A
100
relative intensity
relative intensity
100
80
60
40
80
60
40
20
20
0
0
400
450
500
550
400
wavelength [nm]
450
500
550
600
wavelength [nm]
Figure 3.10: Emission spectra of compounds 38 (A) and 52 (B) [1 µM] in MeCN/0.05% TFA (aq.) at a ratio of 70/30 at two
different excitation wavelengths: 310 nm (blue) and 370 nm (red); Ex Slit: 10, Em Slit: 5.
3.4.3
Alternative fluorescence based ABCG2 inhibition assays
Fluorescent ABC modulators like compounds 51 and 52 (λex: 370 nm; λem: 480 nm) partially overlap
with the excitation and emission spectra of the intercalated Hoechst 33342 dye (λex: 340 nm,
λem: 485 nm). To exclude that the fluorescence of these compounds interferes with the Hoechst
assay, causing false positive results, a microplate assay was established, using mitoxantrone as a
fluorescent ABCG2 substrate with spectral properties different from those of 51 and 52 and that of
intercalated Hoechst 33342. Selected compounds were investigated in the Hoechst 33342, the
mitoxantrone (λex: 612 nm, λem: 670 nm) and additionally in the pheophorbide a (PhA; λex: 380 nm,
λem: 670 nm) microplate assay using MCF-7/Topo cells (Table 3.14).
41
If not otherwise stated, mean values and maximal inhibitory effects were calculated from two to
three independent experiments performed in sextuplicate (Hoechst 33342 assay) or two experiments
performed in triplicate (mitoxantrone and PhA assay).
Table 3.14: Inhibition of ABCG2 by selected modulators, determined in three different microplate
assays using MCF-7/Topo cells. Fluorescent compounds are highlighted in blue.
Compound
12
15
a
H33342 Assay
Mitoxantrone Assay
PhA Assay
λex: 340 nm; λem: 485 nm
IC50 [nM]
Imax (%)
IC50 [nM]
Imax (%)
IC50 [nM]
Imax (%)
591 ± 94
109 ± 8
796 ± 146
94 ± 6
566 ± 74
65 ± 7
89 ± 3
973 ± 87
a
81 ± 2
a
71 ± 3
760 ± 67
98 ± 6
505 ± 13
26
384 ± 47
98 ± 2
275 ± 66
84 ± 2
710 ± 87
29
132 ± 6
121 ± 19
137 ± 4
100 ± 6
488 ± 56 a
146 ± 5
31
160 ± 4
99 ± 1
151 ± 16
101 ± 9
278 ± 22a
125 ± 2
32
59 ± 11
101 ± 5
94 ± 33
99 ± 0
117 ± 34
103 ± 11
33
46 ± 1
111 ± 9
55 ± 13
104 ± 4
101 ± 11
90 ± 2
34
392 ± 40
92 ± 1
627 ± 127
88 ± 5
978 ± 67a
56 ± 1
38
371 ± 90
90 ± 2
426 ± 172
70 ± 6
856 ± 167
53 ± 7
40
145 ± 35
85 ± 6
251 ± 22
90 ± 2
141 ± 21
69 ± 6
51
64 ± 3
95 ± 3
49 ± 6
95 ± 2
480 ± 238
80 ± 2
52
55 ± 14
110 ± 17
115 ± 69
91 ± 12
94 ± 4
89 ± 0
Mean values were calculated from a single experiment performed in triplicate.
As shown in Table 3.14, IC50 values as well as Imax values obtained in the Hoechst 33342 and the
mitoxantrone assay are in a very good accordance. In general, the pheophorbide a assay yielded
lowest activities with the exception of compounds 29, 31 and 32. The fact that these compounds all
belong to the indole series and are substituted at the tetrahydroisochinoline suggests that the
different potencies and efficacies obtained in the PhA assay and the other two ABCG2 modulator
assays are substrate dependent. Besides assay-dependent differences and variable extents of
precipitation of the modulators, the low maximal inhibitory effects in the PhA assay may result from
absorption of the excitation light for PhA by some of the compounds. Figures 3.11-3.13 display the
concentration-response curves for all compounds tested in the Hoechst 33342, the mitoxantrone and
the pheophorbide a microplate assay, respectively.
Whereas compound 51 and 52 yielded almost identical results in the H33342 and the mitoxantrone
assay (λex: 380 nm; λem: 670 nm), compound 38 differs in the efficacy in the two assays.
42
Chapter 3
As shown in Figure 3.13 (B), the inhibition for compound 38 from the mitoxantrone assay does not
deviate from that of the Hoechst assay up to a concentration of 1 µM. As 38 showed very poor
solubility (even in DMSO), it is likely that precipitation was the cause of the discrepancies in the
different ABCG2 assays.
In general, with the establishment of the mitoxantrone assay, a well-suited alternative test system in
terms of ABCG2 inhibition was adjusted. As mitoxantrone fluorescence efflux assays in literature are
mainly performed at the flow cytometer [Matsson et al., 2007], the here arranged assay in the
96-well format represents a comfortable alternative for the screening of a large variety of
compounds.
100
120
B
80
ABCG2 inhibiton (%)
100
A
80
60
40
40
20
20
0
0
0.01
0.1
1
10
concentration [µM]
0.01
100
120
0.1
1
10
concentration [µM]
100
160
C
140
100
60
80
60
40
20
D
120
100
80
60
40
20
0
0
0.01
0.1
1
10
concentration [µM]
100
-20
0.01
0.1
1
concentration [µM]
10
100
Figure 3.11: ABCG2 inhibition by compounds 12 (A), 15 (B), 26 (C) and 29 (D), respectively, in the Hoechst33342 (open
circles), the mitoxantrone (open triangles) and the pheophorbide a (open squares) microplate assay.
120
140
A
100
100
120
43
80
60
40
20
B
80
60
40
20
0
0
-20
0.01
0.1
1
10
concentration [µM]
100
0.01
0.1
1
concentration [µM]
10
100
120
C
100
D
100
120
80
60
40
80
60
40
20
20
0
0
0.01
0.1
1
concentration [µM]
10
0.01
0.1
1
10
100
concentration [µM]
Figure 3.12: ABCG2 inhibition by compounds 31 (A), 32 (B), 33 (C) and 34 (D), respectively, in the Hoechst33342 (open
circles), the mitoxantrone (open triangles) and the pheophorbide a (open squares) microplate assay.
44
Chapter 3
120
100
100
120
A
80
60
40
20
0
0.1
1
concentration [µM]
10
60
40
20
100
0.01
120
0.1
1
10
concentration [µM]
0.01
0.1
concentration [µM]
100
120
C
100
80
0
0.01
100
B
80
60
40
20
D
80
60
40
20
0
0
0.01
0.1
1
concentration [µM]
10
0.001
1
10
Figure 3.13: ABCG2 inhibition by 40 (A) and fluorescent compounds 38 (B), 51 (C) and 52 (D), respectively, in the
Hoechst33342 (open circles), the mitoxantrone (open triangles) and the pheophorbide a (open squares) microplate assay.
3.4.4
Effect of co-administration of ABCG2 inhibitors with
topotecan on the proliferation of MCF-7/Topo cells
To confirm ABCG2 transporter inhibition, resulting in a reversal of ABCG2 mediated drug resistance,
the effect of the five most potent modulators (compounds 32, 33, 50-52) on the proliferation of two
cell lines was investigated using a kinetic crystal violet chemosensitivity assay (Figures 3.14-3.18).
ABCG2-overexpressing MCF-7/Topo cells show resistance against the cytostatic drug topotecan, a
known BCRP substrate, up to concentrations of 550 nM. The cells were incubated with topotecan
alone and in combination with different concentrations of the new modulators. Considering potential
cytotoxicity of the ABCG2 modulators, the cells were also treated with modulators alone.
120
45
2.5
3.0
A
B
120
2.5
2.0
60
1.5
40
1.0
20
2.0
80
T/Ccorr. (%)
80
1.5
60
1.0
(T-Co)/Co (%)
40
absorbance A580
100
absorbance A580
T/Ccorr. (%)
100
0
0.5
0.5
20
-20
-40
0
20
40
60
80
0.0
100 120 140
0
0
20
time of incubation [h]
40
60
80
0.0
100 120 140
Figure 3.14: Effect of compound 32 (UR-COP269) alone (A) and in combination with topotecan (B) on proliferating
MCF-7/Topo cells; vehicle (open stars); A) 32 [1 µM] (filled circles) and [5 µM] (filled triangles), positive control vinblastine
[1 µM] (filled squares); B) topotecan [100 nM] (open circles), topotecan [100 nM] + 32 [100 nM] (filled inverted triangles),
topotecan [100 nM] + 32 [500 nM] (filled diamonds).
120
2.5
120
3.0
A
B
2.0
2.5
80
2.0
60
1.5
1.0
20
40
1.0
20
0.5
absorbance A580
1.5
T/Ccorr. (%)
60
40
(T-Co)/Co (%)
100
80
absorbance A580
T/Ccorr. (%)
100
0
0.5
-20
-40
0.0
0
20
40
60
80
100
0
0.0
0
20
40
60
80
100
120
Figure 3.15: Effect of compound 33 (UR-COP272) alone (A) and in combination with topotecan (B) on proliferating
46
Chapter 3
140
3.0
120
2.0
B
2.5
100
T/Ccorr. (%)
120
A
100
40
1.5
20
1.0
(T-Co)/Co (%)
0
60
1.0
40
absrobance A580
60
80
T/Ccorr. (%)
2.0
absorbance A580
1.5
80
0.5
-20
0.5
20
0.0
80 100 120 140 160
0
-40
-60
0
20
40
60
0
20
40
60
80
0.0
100 120 140
Figure 3.16: Effect of compound 50 (UR-MB107-6) alone (A) and in combination with topotecan (B) on proliferating
120
3.0
120
2.5
A
B
100
100
2.0
80
2.0
40
1.5
20
(T-Co)/Co (%)
1.0
T/Ccorr. (%)
60
absrobance A580
T/Ccorr. (%)
80
1.5
60
1.0
40
absrobance A580
2.5
0
-20
0.5
20
0.0
80 100 120 140 160
0
0.5
-40
0
20
40
60
0
20
40
60
0.0
80 100 120 140 160
3.5
120
100
2.0
2.0
1.5
0
80
T/Ccorr. (%)
50
1.5
60
1.0
40
absorbance A580
2.5
absorbance A580
T/Ccorr. (%)
3.0
(T-Co)/Co (%)
2.5
B
A
100
47
1.0
0.5
0.5
20
-50
0
0.0
20 40 60 80 100 120 140 160 180
0
0
20
40
60
80
0.0
100 120 140
The modulators alone did not affect the proliferation of MCF-7/Topo cells at high concentrations, i.e.
1 µM and 5 µM, respectively. Only compound 32 led to a temporary growth retardation [5 µM]
24 hours after addition of the compound.
The combination of topotecan at a nontoxic concentration of 100 nM with compounds 32 and 33,
respectively, resulted in a cytotoxic drug effect, even at a low modulator concentration of 100 nM. A
higher concentration of the modulator did not increase cyctotoxicity. Compounds 50-52 caused a
strong cytostatic effect on MCF-7/Topo cells in combination with topotecan.
In general, the combination with the cytostatic drug topotecan led to a total reversal of ABCG2
mediated drug resistance by all compounds tested.
As negative control, analogous experiments were performed using U-118 MG cells, which do not
express the ABCG2 transporter. A nontoxic concentration of 10 nM topotecan in combination with
different concentrations of the modulators did not influence cell proliferation at all (as an example
shown for compound 32 in Figure 3.19).
48
Chapter 3
1.6
100
1.4
T/Ccorr. (%)
1.0
60
0.8
0.6
40
absorbance A578
1.2
80
0.4
20
0.2
0
0
20
40
60
80
100
0.0
120
Figure 3.19: Effect of compound 32 (UR-COP269) in
combination with topotecan on proliferating U-118 MG
cells; vehicle (open stars), topo [10 nM] (open circles),
topo [10 nM] + 32 [100 nM] (filled inverted triangles), topo
[10 nM] + 32 [500 nM] (filled diamonds).
3.4.5
Chemical stability of selected ABCG2 transporter modulators
under physiological conditions
As a prerequisite for planned in vivo studies, the most potent inhibitors were subjected to
preliminary biopharmaceutical investigations with a focus on stability in mouse plasma. Because of
the lack of drug like properties (rapid enzymatic cleavage in mouse plasma) of recently described and
characterized 1st and 2nd generation modulators like compounds 1, 2 and 4 [Kühnle, 2010; OchoaPuentes et al., 2011], the chemical structures of these target compounds were modified in a
bioisosteric approach (replacement of the amide) with a main focus on stability resulting in a 3rd and
4th generation of modulators.
Bioisosteric replacements to create new molecules with improved biological properties to the parent
compound are major objectives in successful structure-based drug design. Modification of a
compound can attenuate toxicity and chemical stability, modify activity or alter pharmacokinetics.
Whereas the classical bioisosteres involve structurally simple atoms or groups, nonclassical
bioisosteres replace functional groups with a different number of atoms and/or different steric and
electronic properties [Patani and LaVoie, 1996]. Classical bioisosterism comprises e.g. the
replacement of hydrogen by fluorine classically exemplified by 5-fluoruracil derived from uracil or the
isosteric substitution of an amino group for a hydroxyl group (folic acid and aminopterin).
Nonclassical bioisosterism includes the replacement of noncyclic structures with cyclic residues and
vice versa or the replacement of functional groups [Lemke and Williams, 2012]. A successful story of
bioisosteric replacement in drug discovery was shown for the angiotensin II receptor antagonist
49
losartan: With the replacement of the carboxylic acid residue of its lead structure by a tetrazol the
oral bioavailabilty could significantly be improved [Herr, 2002; Dinges and Lamberth, 2012].
By analogy with the synthesis strategy described in this work, the replacement of a metabolic
instable ester or amide with different heterocycles was shown previously [Eastwood et al., 2011;
Monceaux et al., 2011]. Many other amide and ester isosteres are known including
trifluoroethylamines (successfully shown for the cathepsin K inhibitor odanacatib [Gauthier et al.,
2008]) and oxadiazolines [Sun et al., 2012] as well as isoxazole-ether scaffolds as acetyl group
replacement of acetylcholine [Yu et al., 2012].
3.4.5.1 Chemical stability in culture medium and human CPD plasma
To investigate the stability of the new classes of modulators, the compounds were incubated in
different media, including DMEM supplemented with 10% FCS as well as in murine and human
plasma over a period of 24 hours. HPLC was used for analysis.
Compounds 7-10, 12, 19 and 34, selected representatives of the biphenyl, indole and triazole series,
respectively, were incubated at 37 °C in human plasma or DMEM + 10% FCS. All compounds were
completely stable over a period of 24 hours in both media. Stabilities of compounds 12 and 19 after
24 h are displayed in Figures 3.20-3.21.
250
200
150
A
200
150
signal [mV]
100
signal [mV]
B
50
0
-50
100
50
0
-50
-100
-100
-150
17 18 19 20 21 22 23 24 25 26 27 28
-150
17
retention time [min]
18
19
20
21
22
23
24
25
26
27
28
Figure 3.20: Chromatogram of a DMEM (+ 10% FCS) sample (A) and a human (CPD) plasma sample (B) containing
compound 12 (final concentration: 3 µM) after 24 h of incubation at 37 °C; the respective chromatogram of a blank DMEM
and plasma sample, processed by the same method, is shown as red dotted line.
50
Chapter 3
250
200
200
A
150
B
100
100
signal [mV]
signal [mV]
150
50
0
-50
50
0
-50
-100
-100
-150
-200
17
18
19
20
21
22
23
24
25
26
27
28
-150
17
18
19
20
21
22
23
24
25
26
27
28
Figure 3.21: Chromatogram of a DMEM (+ 10% FCS) sample (A) and a human (CPD) plasma sample (B) containing
compound 19 (final concentration: 3 µM) after 24 h of incubation at 37 °C; the respective chromatogram of a blank DMEM
and plasma sample, processed by the same method, is shown as red dotted line.
3.4.5.2 Chemical stability in mouse plasma
As compounds 1, 2 and 4 are prone to rapid enzymatic cleavage in mouse plasma at the central
benzanilide moiety within 30 min (displayed in Figure 3.22 for compound 4) the new 3rd generation
modulators were also investigated for their chemical stability in murine plasma.
Surprisingly, in mouse plasma, in all compounds of these new series of modulators the ester group is
prone to enzymatic cleavage yielding the inactive carboxylic acid. However, hydrolysis of the ester
occurs much slower than cleavage of the benzanilide of the 1st and 2nd generation modulators.
Regarding stability in murine plasma, all ester bearing indole-, biphenyl- and triazole-type
modulators, respectively, have a half-life of about 24 h compared to 10 min in case of compounds
1, 2 and 4. Additionally, the quinoline carboxamide group was cleaved, though to a very small extent
(as an example: see compound 12 in Figure 3.24).
Although the chemical stability in murine plasma was improved by the renunciation of the lable
amide core, the results are not yet satisfactory. In search for completely stable ABCG2 modulators,
the hydrolysis-sensitive ester group of the 3rd generation triazole series was replaced by a ketone
(compounds 50-52). These ‘4th generation modulators’ were completely stable in murine plasma over
a period of 24 h (Figure 3.26) and are therefore appropriate for in vivo investigations in nude mice.
51
Figure 3.22: Enzymatic degradation of UR-COP78 (4), when incubated in mouse plasma over a period of 24 h at 37 °C:
Complete decomposition to compounds A and B (main cleavage products) within 30 minutes. Additionally, the ester group
and the quinoline carboxamide group were cleaved (HPLC analysis, UV detection at 210 nm). The cleavage products were
identified by HPLC-MS analysis. Fragment E was only detectable by HPLC-MS-analysis (Figure 3.23)
52
Chapter 3
Figure 3.23: HPLC-MS analysis of compound 4 in mouse plasma: Detection of the cleavage products A-E in plasma of NMRI
(nu/nu) mice, 24 h after incubation at 37 °C.
53
Figure 3.24: Enzymatic degradation of UR-COP228 (12), when incubated in mouse plasma over a period of 24 h at 37 °C.
Main cleavage product: compound A, the inactive caboxylic acid after ester hydrolysis. Additional fragments appeard after
degradation of the quinoline carboxamide group (HPLC analysis, UV detection at 210 nm). The cleavage products were
identified by HPLC-MS analysis. Fragment C was only detectable by HPLC-MS-analysis (Figure 3.25)
54
Chapter 3
Figure 3.25: HPLC-MS analysis of compound 12 in mouse plasma: Detection of the cleavage products A-D in plasma of NMRI
(nu/nu) mice, 24 h after incubation at 37 °C.
55
Figure 3.26: Chemical stability of compound 51 when incubated in mouse plasma over a period of 24 h at 37 °C (HPLC
analysis, UV detection at 210 nm).
3.4.5.3 Esterase activity in human and murine plasma
Because of the different stabilities of 3rd generation modulators after incubation for 24 h in human
and murine plasma, respectively, the enzymatic activity of unspecific esterases was compared
spectrophotometrically. Murine plasma showed a 1.5-fold higher esterase activity [6.04 ± 0.20 U/mL]
in comparison to human CPD plasma [4.26 ± 0.26 U/mL] (Stefan Huber, personal communication).
3rd generation modulators bearing an ester group where therefore hydrolyzed in murine, but not in
human plasma. There was no activity difference between CPD and heparinized human blood.
3.4.6
Effect of selected ABCG2 transporter modulators on the
transport of Daunorubicin in a blood-brain barrier model
It was previously shown that the brain multidrug resistance protein (BMDP), the porcine homolog to
the human ABCG2 gene, is highly expressed in porcine brain capillary endothelial cells (PBCEC)
[Eisenblätter and Galla, 2002]. Thus, these cells can be used for transport studies in an established
blood-brain-brain barrier model measuring the capability of BMDP to transport 3H-daunorubicin
across a monolayer of PBCEC grown on porous filters in the presence of ABCG2 inhibitors.
Daunorubicin was previously shown to be a substrate of the human homolog of BMDP [Doyle et al.,
56
Chapter 3
1998]. PBCECs cultured on filter inserts, with the apical chamber corresponding to the blood and the
basolateral chamber corresponding to the brain side in vivo, are morphologically polarized.
Tritium labeled daunorubicin was added to both compartments at identical concentrations. Upon
addition of an ABCG2 inhibitor, the daunorubicin transport across the cell monolayer is hindered. The
amount of daunorubicin which is transported from the basolateral to the apical compartment
inversely correlates with the potency of an ABCG2 modulator.
The selective ABCG2 inhibitors fumitremorgin C (FTC), reference compound for ABCG2 inhibition in
the used fluorescence-based microplate assays, and the two 1st generation modulators UR-ME4-1 (1)
and UR-ME22-1 (2) were used for inhibition studies in the blood-brain barrier (BBB) model
(Figure 3.27). Whereas at a concentration of 1 µM compounds 1 and 2 inhibit the transport of
daunorubicin from the basolateral to the apical side of the PBCEC monolayer almost completely, FTC
180
A
B
8x10-5
160
sucrose permeability [cm/sec]
3
Normalized H-Daunorubicin concentration (%)
produced only moderate inhibition even at a concentration of 10 µM.
140
120
100
80
60
40
6x10-5
4x10-5
2x10-5
0
0
1000
2000
time [min]
3000
4000
0
1000
2000
3000
4000
time [min]
Figure 3.27:
3
A: Transendothelial transport of H-daunorubicin across a PBCEC monolayer in presence of ABCG2 inhibitors:
Fumitremorgin C [10 µM] (squares), compound 1 (UR-ME1-4) [1 µM] (triangles), compound 2 (UR-ME22-1) [1 µM]
(diamonds), control (circles).
3
Concentrations of H-daunorubicin were measured after 0.5, 6, 12, 24, 36, 48 and 72 h in each compartment (apical: blue;
basolateral: red) and normalized to measurements at 0.5 h. The vertical bars for each data point represent the standard
2
deviation of three replicate determinations. Transendothelial electrical resistance (n = 3): 1310 ± 379 Ωcm .
B: Corresponding tightness control of the endothelial cell monolayer in presence of ABCG2 inhibitors indicated by
14
paracellular permeability of C-sucrose.
In the Hoechst 33342 assay, the use of fetal calf serum (FCS) and the contained proteins may be
beneficial for solubility properties of the ABCG2 modulators providing a reservoir in aqueous
systems. Tests in the BBB model, however, were performed under serum-free conditions. Therefore,
it is even more astonishing that compounds 1 and 2 were obviously superior to FTC [10 µM] in the
BBB model at a concentration of 1 µM, whereas in the Hoechst 33342 they only reached 2/3 of the
efficacy of FTC [10 µM] in the presence of FCS. The effective concentrations of free compounds may
be much lower in the BBB model in comparison to the Hoechst 33342 assay.
57
Compounds 32 (UR-COP269) and 33 (UR-COP272) are among the most potent ABCG2 inhibitors
known so far with IC50 values of 59 nM and 46 nM, respectively, and maximal inhibitory effects of
over 100% (Hoechst33342 assay). Although comparable to compounds 1 and 2 in potency and clearly
superior in efficacy (H33342 assay), the extent of daunorubicin transport inhibition was not as
significant as for the aforementioned ABCG2 modulators in the blood-brain barrier model
(Figure 3.28). In general, daunorubicin transport did not significantly change in presence of
compounds 32 and 33, respectively, at a concentration of 1 µM in comparison to the control. A
concentration of 2 µM in case of 32 produced a moderate transport inhibition. The results for 33
[3 µM] are not reliable as sucrose permeability obviously increased when incubating the monolayer
with this high modulator concentration.
Besides different grades of precipitation (water solubility) due to milieu differences between the
Hoechst assay and the BBB model (buffer composition, pH value, hydrocortisone supplementation),
species differences (human ABCG2 transporter in the H33342 assay vs. the porcine homolog in the
1.2x10-4
300
A
250
B
sucrose permeability [cm/sec]
Normalized 3H-Daunorubicin concentration [%]
BBB model) have to be considered for the inconsistent testing results in the two test system.
200
150
100
50
0
-50
10-4
8.0x10-5
6.0x10-5
4.0x10-5
2.0x10-5
0
-100
0
500
1000
1500
2000
time [min]
2500
3000
3500
0
500
1000
1500
2000
2500
3000
3500
time [min]
Figure 3.28:
3
A: Transendothelial transport of H-daunorubicin across a PBCEC monolayer in presence of ABCG2 inhibitors: Compound 32
[1 µM] (filled squares) and [2 µM] (open squares), compound 33 [1 µM] (filled triangles) and [3 µM] (open triangles), control
(circles).
3
Concentrations of H-daunorubicin were measured after 0.5, 6, 12, 24, 36, 48 and 72 h in each compartment (apical: blue;
basolateral: red) and normalized to measurements at 0.5 h. The vertical bars for each data point represent the standard
2
deviation of three replicate determinations. Transendothelial electrical resistance (n = 3): 1617 ± 128 Ωcm .
B: Corresponding tightness control of the endothelial cell monolayer in presence of ABCG2 inhibitors indicated by
14
paracellular permeability of C-sucrose.
To elucidate, whether the synthesized modulators act as substrates or as inhibitors of the ABCG2
transporter, the BBB model experiment was performed as described above with a fixed
concentration of the modulator on both sides of the monolayer. After different periods of time
samples of each compartment were collected and changes in the modulator concentration on the
two sides tried to be calculated by means of HPLC analysis. Unfortunately the concentration of the
compounds in the samples was too low to be significantly quantified by HPLC analysis. The way of
58
Chapter 3
detection was not sensitive enough as only UV detection was available for these classes of
compounds.
3.5
Summary and conclusions
During this work, a series of about 50 new putative ABCG2 transporter modulators was synthesized
in dependence on former described tariquidar analogs [Kühnle, 2010] with a major focus on the
design of more stable, drug-like compounds.
Firstly, the problem of insufficient water solubility of the lead compound UR-ME22-1 (2) was
attempted to be solved by the introduction of triethylenglycol chains resulting in a 2nd generation of
ABCG2 modulators with compound 4 (IC50 130 nM, Imax 88%) as most promising representative. The
insertion of polyethylene glycol moieties led to a clearly increased efficacy.
Secondly, the instable benzamide moiety, point attack of esterases in mouse plasma, was replaced in
a bioisosteric approach giving four new classes of 3rd generation modulators with a biphenyl system,
an indole, a triazole and a quinoline, respectively, as central core structure.
Among these four new classes of BCRP modulators, compounds 29, 32 and 33 bearing an indole are
the most potent ones with IC50 values in the H33342 assay of 132 nM, 59 nM and 46 nM,
respectively, and maximal inhibitory effects of over 100% in comparison to the standard ABCG2
inhibitor fumitremorgin C. Stability tests in mouse plasma revealed that these 3rd generation
modulators are neither immune against enzymatic degradation with the ester as new hydrolysis
weak point.
In search for completely stable ABCG2 modulators, the synthesized structures were further improved
replacing the hydrolysis-sensitive ester by a ketone giving compounds 50-52 which are stable in
mouse plasma over a period of 24 hours. Additionally, together with 32 and 33, compounds 50-52
turned out to be the most potent ABCG2 modulators reported so far.
The results were confirmed in a kinetic chemosensitivity assay where the incubation of topotecan
resistant MCF-7/Topo cells with a combination of topotecan and the modulators led to a complete
reversal of ABCG2-mediated drug resistance.
Because of coincident fluorescent properties of some compounds with the Hoechst 33342 dye, a
second assay system was established using mitoxantrone as fluorescent substrate. Additionally, the
selectivity against other ABC proteins was tested using an already existing calcein-AM assay for tests
at the P-gp [Höcherl, 2010] and a new arranged microplate assay for inhibition tests at the ABCC1
transporter with reversan as standard compound.
59
Inconsistent results were obtained for the testing of compounds 1, 2, 32 and 33 in a porcine bloodbrain barrier model. Although less efficient compared to FTC in the Hoechst 33342 assay, compounds
1 and 2 were clearly superior to the standard compound in the serum-free BBB model. The inhibitory
effect of compounds 32 and 33 in the in vitro model was not as significant as for 1 and 2 although
superior in the cell based microplate assay in both potency and efficacy.
The question rises whether species differences of the expressed ABCG2 transporters contributes to
these results.
Subsequent projects with a major focus on the determination of the way of ABCG2 modulation are of
major interest. The question has to be answered whether the synthesized compounds act as
substrates or as inhibitors at the ABCG2 transporter. Therefore, different approaches are
conceivable, either the use of radio-labeled ABCG2 modulators in the BBB model as a sensitive
method or the testing of the compounds in an ATPase assay (see Chapter 6).
60
3.6
Chapter 3
References
Ahmed-Belkacem, A., Pozza, A., Macalou, S., et al. Inhibitors of cancer cell multidrug resistance
mediated by breast cancer resistance protein (BCRP/ABCG2). Anti-Cancer Drugs 2006, 17(3),
239-243.
Allen, J. D., van Loevezijn, A., Lakhai, J. M., et al. Potent and specific inhibition of the breast cancer
resistance protein multidrug transporter in vitro and in mouse intestine by a novel analogue
of fumitremorgin C. Mol. Cancer. Ther. 2002, 1(6), 417-425.
Bakos, E., Evers, R., Szakacs, G., et al. Functional multidrug resistance protein (MRP1) lacking the Nterminal transmembrane domain. J. Biol. Chem. 1998, 273(48), 32167-32175.
Bates, S. E., Robey, R., Miyake, K., et al. The role of half-transporters in multidrug resistance. J.
Bioenerg. Biomembr. 2001, 33(6), 503-511.
Bauer, S., Ochoa-Puentes, C., Sun, Q., et al. Quinoline Carboxamide-Type ABCG2 Modulators: Indole
and Quinoline Moieties as Anilide Replacements. ChemMedChem 2013, 8, 1773-1778.
Bernhardt, G., Reile, H., Birnböck, H., et al. Standardized kinetic microassay to quantify differential
chemosensitivity on the basis of proliferative activity. J. Cancer Res. Clin. Oncol. 1992, 118(1),
35-43.
Breedveld, P., Beijnen, J. H. and Schellens, J. H. Use of P-glycoprotein and BCRP inhibitors to improve
oral bioavailability and CNS penetration of anticancer drugs. Trends Pharmacol. Sci. 2006,
27(1), 17-24.
Chen, Y. N., Mickley, L. A., Schwartz, A. M., et al. Characterization of adriamycin-resistant human
breast cancer cells which display overexpression of a novel resistance-related membrane
protein. J. Biol. Chem. 1990, 265(17), 10073-10080.
Cooray, H. C., Blackmore, C. G., Maskell, L., et al. Localisation of breast cancer resistance protein in
microvessel endothelium of human brain. Neuroreport 2002, 13(16), 2059-2063.
Crone, C. and Olesen, S. P. Electrical resistance of brain microvascular endothelium. Brain Res. 1982,
241(1), 49-55.
de Bruin, M., Miyake, K., Litman, T., et al. Reversal of resistance by GF120918 in cell lines expressing
the ABC half-transporter, MXR. Cancer Lett. 1999, 146(2), 117-126.
Dinges, J. and Lamberth, C. Bioactive Heterocyclic Compound Classes. Wiley-VCH Verlag GmbH & Co.
KG, Weinheim, Germany, 2012.
Dodic, N., Dumaitre, B., Daugan, A., et al. Synthesis and activity against multidrug resistance in
Chinese hamster ovary cells of new acridone-4-carboxamides. J. Med. Chem. 1995, 38(13),
2418-2426.
Doyle, L. A. and Ross, D. D. Multidrug resistance mediated by the breast cancer resistance protein
BCRP (ABCG2). Oncogene 2003, 22(47), 7340-7358.
Doyle, L. A., Yang, W., Abruzzo, L. V., et al. A multidrug resistance transporter from human MCF-7
breast cancer cells. Proc. Natl. Acad. Sci. U S A 1998, 95(26), 15665-15670.
61
Eastwood, P., Gonzalez, J., Gomez, E., et al. Indolin-2-one p38alpha inhibitors III: bioisosteric amide
replacement. Bioorg. Med. Chem. Lett. 2011, 21(21), 6253-6257.
Egger, M. Inhibition of ABC Transporters Associated with Multidrug Resistance. PhD thesis, University
of Regensburg, Germany, 2009.
Eisenblätter, T. and Galla, H. J. A new multidrug resistance protein at the blood-brain barrier.
Biochem. Biophys. Res. Commun. 2002, 293(4), 1273-1278.
Eisenblätter, T., Hüwel, S. and Galla, H. J. Characterisation of the brain multidrug resistance protein
(BMDP/ABCG2/BCRP) expressed at the blood-brain barrier. Brain Res. 2003, 971(2), 221-231.
Evers, R., Kool, M., van Deemter, L., et al. Drug export activity of the human canalicular multispecific
organic anion transporter in polarized kidney MDCK cells expressing cMOAT (MRP2) cDNA. J.
Clin. Invest. 1998, 101(7), 1310-1319.
Fellner, S., Bauer, B., Miller, D. S., et al. Transport of paclitaxel (Taxol) across the blood-brain barrier
in vitro and in vivo. J. Clin. Invest. 2002, 110(9), 1309-1318.
Fetsch, P. A., Abati, A., Litman, T., et al. Localization of the ABCG2 mitoxantrone resistanceassociated protein in normal tissues. Cancer Lett. 2006, 235(1), 84-92.
Franke, H., Galla, H. and Beuckmann, C. T. Primary cultures of brain microvessel endothelial cells: a
valid and flexible model to study drug transport through the blood-brain barrier in vitro.
Brain Res. Protoc. 2000, 5(3), 248-256.
Franke, H., Galla, H. J. and Beuckmann, C. T. An improved low-permeability in vitro-model of the
blood-brain barrier: transport studies on retinoids, sucrose, haloperidol, caffeine and
mannitol. Brain Res. 1999, 818(1), 65-71.
Gauthier, J. Y., Chauret, N., Cromlish, W., et al. The discovery of odanacatib (MK-0822), a selective
inhibitor of cathepsin K. Bioorg. Med. Chem. Lett. 2008, 18(3), 923-928.
Gottesman, M. M., Fojo, T. and Bates, S. E. Multidrug resistance in cancer: role of ATP-dependent
transporters. Nat. Rev. Cancer 2002, 2(1), 48-58.
Han, B. and Zhang, J. T. Multidrug resistance in cancer chemotherapy and xenobiotic protection
mediated by the half ATP-binding cassette transporter ABCG2. Curr. Med. Chem. Anticancer
Agents 2004, 4(1), 31-42.
Herr, R. J. 5-Substituted-1H-tetrazoles as carboxylic acid isosteres: medicinal chemistry and synthetic
methods. Bioorg. Med. Chem. 2002, 10(11), 3379-3393.
Höcherl, P. New tariquidar-like ABCB1 modulators in cancer chemotherapy: Preclinical
pharmacokinetic / pharmacodynamic investigations and computational studies. PhD thesis,
University of Regensburg, Germany, 2010.
Hoheisel, D., Nitz, T., Franke, H., et al. Hydrocortisone reinforces the blood-brain properties in a
serum free cell culture system. Biochem. Biophys. Res. Commun. 1998, 247(2), 312-315.
Hubensack, M. Approaches to Overcome the Blood-Brain Barrier in the Chemotherapy of Primary
and Secondary Brain Tumors: Modulation of P-glycoprotein 170 and Targeting of the
Transferrin Receptor. PhD thesis, University of Regensburg, Germany, 2005.
62
Chapter 3
Hubensack, M., Müller, C., Höcherl, P., et al. Effect of the ABCB1 modulators elacridar and tariquidar
on the distribution of paclitaxel in nude mice. J. Cancer Res. Clin. Oncol. 2008, 134(5), 597607.
Jäger, W. Classical resistance mechanisms. Int. J. Clin. Pharmacol. Ther. 2009, 47(1), 46-48.
Jekerle, V., Klinkhammer, W., Reilly, R. M., et al. Novel tetrahydroisoquinolin-ethyl-phenylamine
based multidrug resistance inhibitors with broad-spectrum modulating properties. Cancer
Chemother. Pharmacol. 2007, 59(1), 61-69.
Jekerle, V., Klinkhammer, W., Scollard, D. A., et al. In vitro and in vivo evaluation of WK-X-34, a novel
inhibitor of P-glycoprotein and BCRP, using radio imaging techniques. Int. J. Cancer. 2006,
119(2), 414-422.
Kannan, P., Telu, S., Shukla, S., et al. The "Specific" P-Glycoprotein Inhibitor Tariquidar Is Also a
Substrate and an Inhibitor for Breast Cancer Resistance Protein (BCRP/ABCG2). ACS Chem.
Neurosci. 2011, 2(2), 82-89.
Kohno, K., Kikuchi, J., Sato, S., et al. Vincristine-resistant human cancer KB cell line and increased
expression of multidrug-resistance gene. Jpn. J. Cancer Res. 1988, 79(11), 1238-1246.
Kühnle, M., Egger, M., Müller, C., et al. Potent and selective inhibitors of breast cancer resistance
protein (ABCG2) derived from the p-glycoprotein (ABCB1) modulator tariquidar. J. Med.
Chem. 2009, 52(4), 1190-1197.
Lemke, T. L. and Williams, D. A. Foye's Medicinal Chemistry. Lippincott Williams & Wilkin, 2012.
Levin, V. A. Relationship of octanol/water partition coefficient and molecular weight to rat brain
capillary permeability. J. Med. Chem. 1980, 23(6), 682-684.
Matsson, P., Englund, G., Ahlin, G., et al. A global drug inhibition pattern for the human ATP-binding
cassette transporter breast cancer resistance protein (ABCG2). J. Pharmacol. Exp. Ther. 2007,
323(1), 19-30.
Monceaux, C. J., Hirata-Fukae, C., Lam, P. C., et al. Triazole-linked reduced amide isosteres: an
approach for the fragment-based drug discovery of anti-Alzheimer's BACE1 inhibitors. Bioorg.
Med. Chem. Lett. 2011, 21(13), 3992-3996.
Ochoa-Puentes, C., Bauer, S., Kühnle, M., et al. Benzanilide-Biphenyl Replacement: A Bioisosteric
Approach to Quinoline Carboxamide-Type ABCG2 Modulators. ACS Med. Chem. Lett. 2013,
4(4), 393-396.
Med. Chem. Lett. 2011, 21(12), 3654-3657.
Patani, G. A. and LaVoie, E. J. Bioisosterism: A Rational Approach in Drug Design. Chem. Rev. 1996,
96(8), 3147-3176.
63
Rabindran, S. K., Ross, D. D., Doyle, L. A., et al. Fumitremorgin C reverses multidrug resistance in cells
transfected with the breast cancer resistance protein. Cancer Res. 2000, 60(1), 47-50.
Robey, R. W., Polgar, O., Deeken, J., et al. ABCG2: determining its relevance in clinical drug
resistance. Cancer Metastasis Rev. 2007, 26(1), 39-57.
Roe, M., Folkes, A., Ashworth, P., et al. Reversal of P-glycoprotein mediated multidrug resistance by
novel anthranilamide derivatives. Bioorg. Med. Chem. Lett. 1999, 9(4), 595-600.
Sarkadi, B., Ozvegy-Laczka, C., Nemet, K., et al. ABCG2 -- a transporter for all seasons. FEBS. Lett.
2004, 567(1), 116-120.
St-Pierre, M. V., Serrano, M. A., Macias, R. I., et al. Expression of members of the multidrug resistance
protein family in human term placenta. Am. J. Physiol. Regul. Integr. Comp. Physiol. 2000,
279(4), R1495-1503.
Staud, F., Vackova, Z., Pospechova, K., et al. Expression and transport activity of breast cancer
resistance protein (Bcrp/Abcg2) in dually perfused rat placenta and HRP-1 cell line. J.
Pharmacol. Exp. Ther. 2006, 319(1), 53-62.
Sun, Z. Y., Asberom, T., Bara, T., et al. Cyclic Hydroxyamidines as Amide Isosteres: Discovery of
Oxadiazolines and Oxadiazines as Potent and Highly Efficacious gamma-Secretase Modulators
in Vivo. J. Med. Chem. 2012, 55(1), 489-502.
Valdameri, G., Gauthier, C., Terreux, R., et al. Investigation of chalcones as selective inhibitors of the
breast cancer resistance protein: critical role of methoxylation in both inhibition potency and
cytotoxicity. J. Med. Chem. 2012, 55(7), 3193-3200.
Valdameri, G., Genoux-Bastide, E., Peres, B., et al. Substituted chromones as highly potent nontoxic
inhibitors, specific for the breast cancer resistance protein. J. Med. Chem. 2012, 55(2), 966970.
Valdameri, G., Pereira Rangel, L., Spatafora, C., et al. Methoxy stilbenes as potent, specific,
untransported, and noncytotoxic inhibitors of breast cancer resistance protein. ACS Chem.
Biol. 2012, 7(2), 322-330.
Wegener, J., Hakvoort, A. and Galla, H. J. Barrier function of porcine choroid plexus epithelial cells is
modulated by cAMP-dependent pathways in vitro. Brain Res. 2000, 853(1), 115-124.
Winter, E., Lecerf-Schmidt, F., Gozzi, G., et al. Structure-Activity Relationships of Chromone
Derivatives toward the Mechanism of Interaction with and Inhibition of Breast Cancer
Resistance Protein ABCG2. J. Med. Chem. 2013, 56(24), 9849-9860.
Yamazaki, R., Nishiyama, Y., Furuta, T., et al. Novel acrylonitrile derivatives, YHO-13177 and YHO13351, reverse BCRP/ABCG2-mediated drug resistance in vitro and in vivo. Mol. Cancer. Ther.
2011, 10(7), 1252-1263.
Yu, L. F., Tuckmantel, W., Eaton, J. B., et al. Identification of novel alpha4beta2-nicotinic acetylcholine
receptor (nAChR) agonists based on an isoxazole ether scaffold that demonstrate
antidepressant-like activity. J. Med. Chem. 2012, 55(2), 812-823.
Zhang, S., Wang, X., Sagawa, K., et al. Flavonoids chrysin and benzoflavone, potent breast cancer
resistance protein inhibitors, have no significant effect on topotecan pharmacokinetics in rats
or mdr1a/1b (-/-) mice. Drug. Metab. Dispos. 2005, 33(3), 341-348.
64
Chapter 3
Zhang, S., Yang, X. and Morris, M. E. Flavonoids are inhibitors of breast cancer resistance protein
(ABCG2)-mediated transport. Mol. Pharmacol. 2004, 65(5), 1208-1216.
Chapter 4
4
Drug-like properties of new
ABCG2 modulators
4.1
Introduction
4.1.1
Trends in medicinal chemistry and drug development
Nowadays, pharmaceutical companies mainly perform high-throughput screening to discover so
called ‘hits’, which are subject of hit-to-lead processes to identify lead compounds for further
optimization, aiming at the development of potential drug candidates. Computational methods as
well as automated sample preparation and screening procedures led to drug libraries with an
enormous number of new compounds. These techniques are used to improve chemical structures
regarding biological activity to predict the compounds’ physicochemical properties as well as drug
absorption or oral bioavailability [Stegemann et al., 2007]. However, improving a compound’s
potency at the cost of solubility and bioavailability is a prominent negative trend in ‘rational drug
design’ mainly resulting from higher H-bonding properties and increasing molecular weights
attended by a enhanced lipophilicity, [Lipinski, 2000]. As a consequence, numerous compounds
never pass clinical trials or become commercially available drugs.
4.1.2
The ‘Rule of Five’
Lipinski’s ‘Rule of 5’ is based on the drug-like properties of several thousand drugs [Lipinski et al.,
2001] and a useful fingerpost or starting point in drug design and discovery with regard to
bioavailability.
66
Chapter 4
According to this rule, good absorption and permeation is more likely when:
1. There are not more than 5 H-bond donors (sum of OHs and NHs);
2. There are not more than 10 H-bond acceptors (sum of Os and Ns);
3. The molecular weight is ≤ 500 Da;
4. The logP is ≤ 5.
The ‘Rule of 5’ addresses all therapeutic areas. Nonetheless, there are occasional limitations. It
should be modified for centrally active compounds [Pajouhesh and Lenz, 2005; Reichel, 2006] as CNS
penetration is more likely if:
1. There are ≤ 3 H-bond donors;
2. There are ≤ 7 H-bond acceptors;
3. The molecular weight is < 400-500 Da;
4. The logP is ≤ 5.
Furthermore, it is limited, if there is a biological transporter involved. The commonly applied
molecular weight cutoff of 500 Da per se, is not an excluding criterion regarding acceptable
bioavailability [Veber et al., 2002], as shown for successful big drug molecules such as cyclosporine
[Ptachcinski et al., 1986] or tacrolimus (FK506) [Venkataramanan et al., 1995].
4.1.3
Plasma protein binding
Besides the fulfillment of various physicochemical requirements, information about plasma protein
binding (PPB) of drug candidates is key in drug development to enable estimates and extrapolations
to in vivo kinetics as early as possible. Different methods for quantifying the binding of drugs to
plasma proteins are available: equilibrium dialysis, ultrafiltration and ultracentrifugation are the most
common ones. The results are usually expressed as %PPB (equation 4), where [D] is the free drug
concentration and [DP] is the concentration of the drug protein complex:
(e ua on 4)
Based on binding experiments, actual concentrations of the free drug are calculated and used as a
base for in vivo experiments. Information on PPB is also of great importance with regard to toxicity,
because plasma proteins may act as reservoirs, and accumulation of drugs in tissues may
substantially influence their plasma concentrations and retention times in the body.
Drug-like properties of new ABCG2 modulators
67
The extent of drug distribution into tissues mainly depends on the degree of binding to the plasma
proteins albumin, alpha-1-acid glycoprotein and lipoproteins. As only the free (unbound) drug is
available for distribution in the body to get access to cellular targets, extensive PPB limits a drug’s
efficacy and influences its pharmacokinetics. On the other hand, drugs (mostly with relatively poor
solubility), are often more easily transported in vivo when bound to plasma proteins.
Prominent clinical drugs with a high protein binding of over 90% are the antieptileptic drug
phenytoin [Lund et al., 1972; Peterson et al., 1982], the anticoagulant warfarin [Mungall et al., 1984]
or the anti-diabetic agent tolbutamide [Miners et al., 1982].
A further phenomenon observed in binding studies, is displacement of the drug from protein:
interactions between two compounds can increase the free fraction of a drug highly bound to plasma
proteins as the interacting drug displaces the substance of interest from its binding site. Drug
displacement from PPB has to be regarded with suspicion due to potential clinically relevant
interactions [Rolan, 1994] including unanticipated cross-reactions in terms of toxicity.
The phenomenon of poor water solubility and high lipophilicity is often associated with increased
binding to proteins. However, for targeting the CNS, lipophilicity of compounds is favorable to
overcome the endothelium of the blood-brain barrier [Waterhouse, 2003].
The discrepancies between solubility, plasma protein binding and BBB penetration of drug
candidates pose a challenge even in case of ‘drug-like’ compounds, as the physico-chemical
properties are not always predictive for adequate access to the brain and a corresponding biological
activity in the CNS [Abbott et al., 2010].
To find candidates for future in vivo studies in nude mice, the characterization of the most potent
ABCG2 modulators with respect to solubility and plasma protein binding is required. Attempts are
described in this chapter.
68
Chapter 4
4.2
4.2.1
Drugs and chemicals
PBS was made of 8.0 g/L NaCl, 1.0 g/L Na2HPO4 · 2 H2O, 0.20 g/L KCl, 0.20 g/L KH2PO4 and 0.15 g/L
NaH2PO4 · H2O (pH 7.4). Fetal calf serum (FCS) was obtained from Biochrom (Berlin, Germany) and
Hoechst 33342 from Invitrogen (Karlsruhe, Germany). Fluorescence was measured with a LS50 B
spectrometer from Perkin Elmer (Rodgau, Germany). If not otherwise stated, chemicals (p.a. quality)
were obtained from Merck (Darmstadt, Germany). Milli-Q system (Millipore, Eschborn, Germany)
was used for water purification.
4.2.2
Solubility
4.2.2.1 Comparison of solubilities in different media
As poor water solubility is a characteristic feature of the classes of ABCG2 modulators synthesized in
our group, the solubility of representative examples, compounds 32, 33, 51, and 52, in DMSO was
compared to the solubility in PBS (pH 7.4), determined by HPLC. Therefore, the same concentrations
of the compound (0.5-20 µM) were prepared in DMSO and PBS, respectively. The samples were
centrifuged (10 min, 16,000 g) with an Eppendorf centrifuge 5415R (Eppendorf, Hamburg, Germany)
to separate undissolved compound, and the supernatants were filtered through Phenex™-RC 4 mm
Syringe Filters 0.2 µm (Phenomenex, Aschaffenbug, Germany). As the compounds are readily soluble
in DMSO, a linear calibration curve was calculated from the peak areas obtained for the samples
diluted with DMSO, assuming that the compounds were completely dissolved. The peak areas for the
corresponding centrifuged PBS samples were referred to the calibration line to calculate solute
concentrations in buffer.
UV detection of compounds 32, 51 and 52 was performed at 210 nm with a Waters (Eschborn,
Germany) system using a 2487 UV detector (described in section 3.3.7.3) or a Merck-Hitachi HPLC
system composed of a L-5000 controller, a 655A-12 pump, a 655A-40 autosampler, a F-1050
fluorescence spectrometer and a L-4250 UV-VIS detector on a Eurospher-100 C18 column
(250 × 4 mm, 5 μm; Knauer, Berlin, Germany) at a flow rate of 0.8 mL/min under helium degassing
and the following gradient: MeCN/0.05% TFA (aq.): 0 min: 60/40, 30 min: 95/5, 31 min: 95/5, 33 min:
95/5. UV detection for compound 33 was performed at 280 nm and fluorescence detection for
compound 51 at 470 nm after excitation at 370 nm.
69
4.2.2.2 Fluorescence spectra of Hoechst 33342 after intercalation in DNA in
the absence and presence of test compounds
All ABC transporter inhibition assays, described in Chapter 3, were performed either in culture
medium supplemented with FCS or in buffer containing serum albumin. The question arose, why the
newly synthesized compounds are so potent and efficient in these aqueous assay systems despite
obviously limited water solubility. In this context it is of utmost importance to ensure that the
measured increase in fluorescence in the assay, resulted from ABC transporter modulation and
subsequent DNA intercalation of the dye (Hoechst 33342), but not from, cell-dependent interference
between the Hoechst dye and the compounds. Therefore, compounds 33, 34 and 51 were incubated
without MCF-7/Topo cells in the presence and the absence of Hoechst 33342 on the basis of the
protocol of the ABCG2 inhibition assay: the compounds were diluted in PBS containing 10% FCS and
8 µM of the Hoechst 33342 dye to a final concentration of 10 µM according to the standard
procedure described in section 3.3.5.1. DNA (10 µg/mL), isolated from calf thymus, was added to the
mixture to allow a DNA intercalation. DMSO served as negative control. The fluorescence spectrum
of Hoechst 33342 in the presence of 1% (v/v) DMSO was compared with the spectra in the presence
of the modulators.
4.2.3
Plasma protein binding studies
The binding of hydrophobic molecules to human serum albumin (HSA) or bovine serum albumin
(BSA), an ortholog of HSA, has been extensively studied for years. Albumins are the most abundant
proteins in the blood circulation (55% of all serum proteins) [Anderson and Anderson, 2002] and
known to act as transporters for various drugs [Takehara et al., 2009]. To determine the extent of
plasma protein binding of the newly synthesized modulators, the applicability of different methods
was explored.
4.2.3.1 Ultrafiltration
Ultrafiltration was performed with Amicon Ultra-0.5 mL Centrifugal Filters (regenerated cellulose
membrane, nominal molecular weight limit: 10 kDa) from Merck Millipore (Darmstadt, Germany) and
Nanosep® Centrifugal Devices (modified polyethersulfone membrane) from Pall GmbH (Dreieich,
Germany), respectively, at 16,000 g using an Eppendorf 5415R centrifuge.
70
Chapter 4
4.2.3.2 Isothermal titration calorimetry
Isothermal titration calorimetry (ITC) is a widely used technique to study biological processes at the
molecular level, especially to characterize binding affinities of ligands to a protein. Thereby, ITC
measures small changes of temperature during the reaction and allows the simultaneous
determination of thermodynamic parameters such as heat of binding (enthalpy; ∆H), number of
binding sites (n) or the calculation of the binding constant K [Santos et al., 2007]. ITC experiments
and the instrumental set-up (Figure 4.1) are described below.
The solution in the syringe containing the test compound
was titrated into the sample cell (37 °C) with the protein
Syringe
solution. Every injection causes a change in heat, resulting in
a change in temperature in the sample cell in comparison to
the reference cell. The change in power, which was required
to return the sample cell to 37 °C, was recorded with every
injection till all binding sites of the protein in the sample cell
were saturated (heat signal diminishes).
Isothermal titration calorimetry was carried out at the
Institute of Organic Chemistry (Prof. Dr. B. König) using a
ΔT
VP-ITC Micro Calorimeter (MicroCal, Northhampton, MA).
Selected synthesized modulators were used as titrants and
were pipetted into a BSA solution (sample cell). To ensure
that the established method and instrumental set-up
worked, a standard compound was tested: propranolol
hydrochloride (Sigma, Munich, Germany), soluble in water,
was chosen as test compound because corresponding
binding constants are extensively referenced in the
literature. Therefore, a solution of 5 mM propranolol HCl (in
PBS) was injected in 5 µL-steps into the sample cell
containing a 1 mM protein (BSA) solution (in PBS).
Reference
Cell
Sample
Cell
Figure 4.1: An ITC instrument consists of
two identical cells surrounded by an
adiabatic jacket (yellow). Reference heater
(green), cell feedback heater (black) and
calibration heater (red) detect temperature
differences between the reference and the
sample cell upon ligand titration into the
sample cell containing the protein;
illustration adapted from [Pierce et al.,
1999].
4.2.3.3 Equilibrium dialysis
To evaluate the extent of protein binding, equilibrium dialysis was performed with DispoEquilibrium
DIALYZERS and a Fast Micro-Equilibrium DIALYZER dialysis chamber (250 µL chamber; polycarbonate
membrane, 0.01 µm pore size) from Harvard Apparatus (Holliston, MA, USA).
A
B
71
C
Figure 4.2: A, B) Fast Micro-Equilibrium DIALYZER dialysis chamber (250 µL); C) DispoEquilibrium DIALYZER; Harvard
Apparatus (Holliston, MA).
The theory behind is as follows (according to ‘Guide to Equilibrium Dialysis’, Harvard Apparatus;
http://www.nestgrp.com/pdf/Ap1/EqDialManual.pdf): two chambers are separated by a dialysis
membrane with a defined molecular weight cut off (MWCO) such that the protein-bound compound
is retained. A known concentration of a compound - small enough to pass freely through the
membrane - is placed into one of the chambers (total volume: 250 µL). An equivalent volume of
protein (BSA), dissolved in the same solvent at defined concentration is then placed in the other
chamber. The compound diffuses across the membrane and either binds to the protein to a certain
extent or remains free in solution. Diffusion across the membrane and binding continues until
equilibrium has been reached. At equilibrium, the concentration of the free compound in solution is
the same in both chambers and can be used to evaluate the binding characteristics of the sample.
To consider that the used compounds show sufficient solubility and cross the membrane, recovery
studies were performed without protein with different synthesized modulators and quantified by
HPLC. Recovery analysis was performed at 37 °C (water bath) under stirring for 4 h or overnight. For
this purpose, compounds 4, 50, 51 and 52, respectively, were dissolved at a concentration of 100 µM
in PBS or PBS supplemented with various amounts of DMSO (5%, 10%, 15% or 20%).
4.2.3.4 Protein binding analysis via fluorescent spectroscopy
Fluorescence measurements were performed at the LS50 B spectrometer from Perkin Elmer at a scan
rate of 240. Hellma® fluorescence cuvettes (SUPRASIL® quartz, semi Micro, pathlength 10 mm) and
fatty acid/globulin free bovine serum albumin (BSA; ≥98% purity) were both purchased from Sigma
(Munich, Germany). Fluorescence spectra were recorded after excitation at 280 nm (BSA) or 370 nm
(fluorescent ABCG2 modulators).
72
Chapter 4
4.2.3.5 HPLC-based methods to determine protein binding of selected
ABCG2 modulators
Three representative ABCG2 modulators were kindly characterized in the laboratory of Origenis
GmbH (Martinsried, Germany) by means of three different methods to estimate the extent of plasma
protein binding.
The chromatographic hydrophobicity index (CHI) was calculated using a C18 column by converting
retention times to volume percentage organic-phase concentrations. The CHI scale can be related to
octanol/water partition coefficients (logD values at pH 7.4) by taking the number of H-bond donor
groups in the molecules into account.
Binding was measured on a column containing immobilized HSA, and membrane affinity was
calculated via immobilized artifical membrane (IAM) HPLC analysis to model drug partitioning
between an aqueous (mobile) phase and a cell membrane (IAM column). By all three methods,
basically, the retention times of the compounds were measured (UV/VIS detection) and linearly
(y = a·Tr + b) converted into the corresponding parameter based on a calibration data set of
thousands of compounds (Dr. M. Thormann, personal communication, 2013).
4.3
4.3.1
Solubility of selected ABCG2 modulators
73
All investigations described in this chapter were severely compromised by poor water solubility of
the synthesized ABCG2 modulators. To obtain at least a rough estimation of lipophilicity and
solubility in water, different methods were applied.
4.3.1.1 Computational calculations
First calculations of the lipophilicity of selected ABCG2 modulators were performed with the
ACD/Labs chemistry software 12.0 (Advanced Chemistry Development, Inc., Toronto, Canada).
Table 4.1: Computational predictions regarding water solubility of selected ABCG2 modulators.
Compd.
Molecular Weight
logP a
logD b
[Da]
a
Solubility c
[mg/L]
4
776.87
4.65
4.12
0.12
12
719.82
4.90
4.66
0.36
16
866.01
4.45
3.91
9.49
31
640.73
5.68
5.08
0.04
32
772.88
4.91
4.32
0. 44
33
891.02
3.66
3.08
14.00
40
654.71
4.98
4.56
0.95
50
652.74
4.90
4.48
1.28
51
666.77
5.42
5.05
0.47
52
798.93
4.65
4.28
4.12
42
770.87
4.76
4.69
0.08
47
721.84
5.75
3.52
1.97
Octanol-water partition coefficient: ratio of a compound’s concentration in octanol and
b
water at equilibrium, expressed as its decadic logarithm logP; distribution constant at
c
pH 7.4; solubility in water at pH 7.4.
74
Chapter 4
4.3.1.2 Solubility in DMSO and PBS
Compounds 32, 33, 51, and 52 were investigated for their solubility in PBS and DMSO by means of
HPLC analysis. Compounds 32, 33 and 51 were neither detectable by UV absorbance in the
supernatant of the centrifuged PBS samples at 210 nm (Waters HPLC system) or 280 nm (compound
33; Merck HPLC system), respectively, nor by fluorescence detection (compound 51; Merck HPLC
system). Although the detection limit of the used UV detector and different spectral properties of the
modulators have to be considered, these negative results reveal that compounds 32, 33 and 51 are
virtually insoluble in aqueous media. By contrast, compound 52 was detectable in PBS though at a
very low concentration (Figure 4.3).
10000
Area [mV*sec]
8000
Figure 4.3: Main peak areas of
compound 52 when diluted in DMSO
(circles) or PBS (squares) at various
concentrations obtained after RP-HPLC
analysis performed with a Waters
(Eschborn, Germany) HPLC system (UV
detection at 210 nm); the applied
gradient was as following: MeCN/0.05%
TFA (aq.): 0 min: 15/85, 25 min: 80/20,
26 min: 95/5, 36 min: 95/5, 37 min:
15/85, 45 min: 15/85 at a constant flow
rate of 1.0 mL/min.
6000
4000
2000
0
0
5
10
15
20
concentration [µM]
The peak areas in the chromatograms of compound 52 for the centrifuged PBS samples were
equivalent to the following free concentrations in buffer:
applied
concentration
[µM]
found
concentration
[µM]
0.50
-
1
0.04
5
0.34
10
0.60
20
0.53
Table 4.2: Found concentrations of compound 52 in PBS after
centrifugation.
The highest concentration determined for compound 52 (0.6 µM) corresponds to a solubility of
0.48 mg/L in PBS. This value is ten times lower than the solubility calculated with ACD/Labs.
75
4.3.1.3 Fluorescence spectra of the Hoechst 33342 dye in the presence of
different modulators
The fluorescence intensity of the Hoechst 33342 dye in presence of a modulator does not differ from
the negative control (H33342 + DMSO) (Figure 4.4).
Fluorescence of the Hoechst 33342 dye is not affected by the combination with different modulators
under Hoechst assay conditions, suggesting that the measured increase in fluorescence intensities in
the Hoechst 33342 microplate assay, indeed, results from ABCG2 inhibition and subsequent DNA
intercalation.
Hence, the modulators work in an aqueous assay system, despite low water solubility. It may be
speculated that the synthesized compounds are actually much more potent than observed, i.e. lower
concentrations suffice to reach a high ABCG2 inhibition, or they bind to protein in the medium
(supplemented with FCS) and are thus brought into solution.
180
160
160
A
140
120
120
relative intensity
relative intensity
140
B
100
80
60
40
100
80
60
40
20
20
0
0
-20
400
450
500
550
600
650
wavelength [nm]
-20
400
450
500
550
600
650
wavelength [nm]
180
160
C
relative intensity
140
120
Figure 4.4: Fluorescence spectra of the Hoechst 33342
dye [8 µM] in PBS containing 10% FCS and 10 µg/mL
DNA in presence of DMSO (black line) or compound
33 (A), 34 (B) and 51 (C) (red) performed at a LS50 B
spectrofluorimeter from Perkin Elmer (Rodgau,
Germany). The corresponding compounds diluted in
PBS/FCS without H33342 are displayed in blue.
Instrument settings were as follows: λex: 340 nm;
Ex/Em Slit: 10.
100
80
60
40
20
0
-20
400
450
500
550
wavelength [nm]
600
650
76
Chapter 4
4.3.2
Extent of plasma protein binding
4.3.2.1 Ultrafiltration and ITC
Determination of plasma protein binding by ultrafiltration failed. Recovery studies revealed that the
tested compounds completely adsorbed to the cellulose membranes.
Isothermal titration calorimetry was successfully performed for the test compound propranolol HCl
(cf. Figure 4.5).
10
µcal/sec
8
6
4
Figure 4.5: Titration thermogram of raw
injection heats for titration of propranolol
HCl [5 mM] into a BSA solution [1 mM].
2
0
20
40
60
80
100
120
140
160
180
200
Time (min)
In contrast, binding experiments via ITC were not successful with the synthesized ABCG2 modulators
as these compounds are almost insoluble in water, and the composition of the solvent should be
exactly the same for the titrant and the cell solution. Blank-experiments (negative control) with
DMSO (titrant) as appropriate solvent for the synthesized compounds and BSA (in H20; sample cell)
were performed prior to experiments, but produced unreliable results when the organic solvent was
mixed with water. As DMSO causes protein precipitation and denaturation, pure DMSO is
inappropriate as solvent for both titrant and BSA.
4.3.2.2 Equilibrium dialysis
Prior to PPB studies via equilibrium dialysis, recovery tests were performed with selected
modulators. Compounds 4, 50, 51 and 52 were dissolved in PBS supplemented with DMSO (5%, 10%,
15% and 20%, respectively) and added to the sample chamber of the Fast Micro-Equilibrium
DIALYZER. Unfortunately, after incubation over night at 37 °C, only compound 4 could be recovered
to a small extent in PBS/DMSO at a ratio of at least 85/15 (HPLC analysis, UV detection at 210 nm),
though, inadequate for PPB tests.
77
In general, all tested compounds were insoluble in pure PBS buffer. Likewise, DispoEquilibrium
DIALYZERS were inappropriate for dialysis experiments as all tested compounds were adsorbed to
the regenerated cellulose membrane.
4.3.2.3 Investigations on fluorescent modulators to determine protein
binding
Intrinsic fluorescence of proteins is mainly due to the aromatic amino acids tryptophan (Trp) and
(Tyr) (Figure 4.6), whereas the indole groups of Trp residues are the dominant source of emission.
Figure 4.6: Chemical
structures of the
aromatic
amino
acids
tryptophan
and tyrosine.
Protein-protein association, protein unfolding as well as the binding of a ligand to a protein alters the
sensitive microenvironment of Trp and its particular location within the protein thereby causing
changes in the protein’s tertiary and quaternary structure, resulting in changes in the fluorescence
spectrum like maximum shifts to the red or the blue as well as intensity changes [Vivian and Callis,
2001].
Bovine serum albumin (66.2 kDa) contains 21 tyrosine and 2 tryptophan residues [Peters, 1985] that
possess intrinsic fluorescence with Trp212, located in the core of the protein and Trp134 found on
the surface of albumin. HSA, which shares 76% homology to BSA, contains only one Trp214 residue
[Moriyama et al., 1996; Bujacz, 2012; Majorek et al., 2012; Patra et al., 2012].
Selected compounds of the triazole series described in Chapter 3, show distinctive fluorescence in
organic solvents like chloroform (Figure 4.7). As fluorescent spectroscopy in protein analysis for
investigations concerning protein binding and protein-ligand interactions is widely described in
literature [Moriyama et al., 1996; Ladokhin, 2000; Takehara et al., 2009; Galecki et al., 2012],
fluorescence may be used to determine the binding of the synthesized ABCG2 modulators to serum
albumin.
78
Chapter 4
relative Intensity
600000
Figure 4.7: Emission spectrum of compound
34 in chloroform after excitation at 380 nm
recorded at a Horiba ‘Fluoromax 4’
fluorescence spectrophotometer (HORIBA
Jobin Yvon GmbH, Unterhaching, Germany)
with Hellma® quartz cuvettes (1 cm); slit
width: 10 nm.
400000
200000
0
400
450
500
550
600
Wavelength (nm)
Changes in either the fluorescence of the protein or the compounds upon binding can be exploited to
study protein-ligand interactions. With excitation wavelengths between 370 and 380 nm and an
emission around 470 nm, the fluorescence properties of the compounds do not interfere with those
of serum albumin (λex: 280 nm; λem: 350 nm).
Compound 52, which is insoluble and therefore non-fluorescent in PBS (pH 7.4), shows weak
fluorescence in DMSO (Figure 4.8). As described in Chapter 3, selected compounds show
fluorescence under acidic HPLC conditions. With the addition of trifluoroacetic acid (TFA; Sigma,
Munich, Germany) to the DMSO samples, the fluorescence intensity of 52 significantly increased (for
further details also cf. section 4.3.3).
350
relative intensity
300
250
200
Figure 4.8: Fluorescence spectra of compound 52
[1 µM] after excitation at 370 nm in PBS (pH 7.4)
(black dotted line), DMSO (black) and DMSO/TFA
(aq.) at pH 5 (blue) and pH 3 (red); Ex/Em Slit: 5.
150
100
50
0
400
450
500
550
600
wavelength [nm]
Due to insufficient solubility of fluorescent compounds in aqueous media, a minimum amount of 15%
DMSO is required for spectroscopy. The pH dependent fluorescence of compound 52 at a
concentration of 1 µM in PBS/DMSO (85/15) is shown in Figure 4.9.
79
250
relative intensity
200
150
Figure 4.9: Fluorescence spectra of 52 [1 µM] in
PBS/DMSO 85/15 at pH 6.5 (gray), pH 6.0 (green),
pH 5.5 (orange), pH 5.0 (blue) and pH 3.0 (red),
respectively; λex: 370 nm; Ex/Em Slit: 10.
100
50
0
400
450
500
550
600
wavelength [nm]
As fluorescence of albumin is affected by the local environment and thereby depending upon the
solvent polarity, the effect of DMSO on the structural conformation of bovine serum albumin was
estimated by monitoring the fluorescence properties of the protein.
Fluorescence spectra of BSA (λex: 280 nm) dissolved in different PBS/DMSO combinations are shown
in Figure 4.10: The fluorescence intensity decreased with increasing amounts of DMSO and a slight
shift of the emission maximum from 350 nm (PBS) to 335 nm (PBS/DMSO 1:1) was visible.
700
600
realtive intensity
500
Figure 4.10: Fluorescence spectra of BSA
[0.05 mg/mL] after excitation at 280 nm in different
media: PBS (black) or PBS/DMSO: 95/5 (dark blue),
90/10 (blue), 85/15 (pink), 80/20 (green), 75/25
(orange), 70/30 (gray), 50/50 (red); Ex/Em Slit: 5.
400
300
200
100
0
300
350
400
450
500
wavelength [nm]
In general, two contrary effects have to be considered influencing the fluorescent properties of
albumin: by raising the percentage of DMSO, on the one hand, the polarity of the microenvironment
of the protein (especially of the aromatic amino acid tryptophan) is drastically changed, and, on the
other hand, an increasing DMSO concentration is assumed to denature and thereby unfold the
protein. Precipitation also contributes to the decrease in fluorescence intensity so that a linear
correlation between fluorescence intensities and the DMSO content cannot be hypothesized
especially at a high DMSO content of 30% or 50%.
80
Chapter 4
Similar results were recently published by Pabbathi et al. [Pabbathi et al., 2013]. The study revealed a
doubling of the hydrodynamic radius of BSA in presence of 40% DMSO, suggesting complete
unfolding of the protein.
A required amount of 15% DMSO due to solubility problems of ABCG2 modulators, indeed, hardly
influenced the fluorescence properties of BSA (cf. Figure 4.10) suggesting that the native structure of
BSA was retained under these conditions.
However, the attempt to quantify the extent of protein binding to BSA of the fluorescent compounds
51 and 52 in a mixture of PBS and DMSO at a ratio of 85:15 at pH 7 failed, due to lack of fluorescence
emission at neutral pH. To guarantee sufficient fluorescence intensity of compounds 51 and 52,
protein binding investigations by means of fluorescent measurement have to be performed at pH
values below 6.5 (Figure 4.9).
Moreover, the isoelectric point (IEP) of BSA is at pH 4.7 [Ge et al., 1998]. At pH 5 and below, acidinduced unfolding and denaturation of the protein occurs [Estey et al., 2006]. As shown in
Figure 4.11, the fluorescence intensities of BSA are decreasing from pH 5.5 to pH 3 as a result of
denaturation.
700
700
B
600
650
500
600
relative intensity
relative intensity
A
400
300
200
500
450
400
100
0
300
550
350
400
wavelength [nm]
450
500
350
7.0
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
pH value
Figure 4.11: A) Fluorescence spectra of BSA [0.01 mg/mL] after excitation at 280 nm in PBS (pH 7; black dotted line) and in
PBS/DMSO (85/15) at pH 7.0 (black), pH 6.5 (gray), pH 6.0 (green), pH 5.5 (orange), pH 5.0 (blue) and pH 3.0 (red),
respectively; B) Maximal fluorescence intensities of BSA in PBS/DMSO 85/15 at 350 nm at descreasing pH value;
Ex/Em Slit: 10.
Based on these results, a pH of 6.0 and a combination of PBS/DMSO at a ratio of 85/15 seems to be
appropriate for fluorescence based binding experiments with compounds 51 and 52, performed at
the excitation and emission wavelengths of the modulators as sensitive changes in the fluorescence
properties of BSA due to the use of DMSO may be mistaken for binding effects. Figure 4.12 shows
the fluorescence emission of compound 52 [1 µM] in the presence of bovine serum albumin at a
molar ratio of 1/1.5 and 1/15 (BSA/ligand), respectively, over 60 minutes. The addition of BSA is
81
accompanied by an increase in the quantum yield of the fluorescent compound. Afterwards, the
intensity continuously decreases within 30 minutes followed by a slight increase after 45 min to a
constant value. The fluorescence intensity of compound 52 in presence of 0.1 mg/mL BSA halved
within 1 h, whereas in presence of 1.0 mg/mL it was diminished to 70%. The effect of an increased
quantum yield did not change significantly with different protein concentrations added.
350
350
A
300
250
relative intensity
realtive intensity
300
200
150
250
200
150
100
100
50
50
0
400
450
500
550
600
B
0
400
450
500
550
600
wavelength [nm]
wavelength [nm]
350
realitve intensity at 470 nm
C
300
250
200
150
100
0
10
20
30
40
50
60
min after BSA addition
Figure 4.12: A,B) Fluorescence spectra of compound 52 [1 µM] after excitation at 370 nm in PBS/DMSO (85/15) at pH 6.0
alone (black dotted line) and in the same solvent in the presence of BSA [0.1 mg/mL] (A) or [1.0 mg/mL] (B) at different
periods of time after BSA addition: 0 min (black), 5 min (red), 10 min (green), 20 min (orange), 30 min (gray), 45 min (blue),
60 min (dark blue); C) Maximal fluorescence intensities of 52 at 470 nm in PBS/DMSO containing 0.1 mg/mL (filled circles)
and 1.0 mg/mL BSA (filled trianlges), respectively, as a function of time after protein addition; Ex/Em Slit: 10.
The gradient in fluorescence intensity over 60 min is difficult to interpret. It is conceivable that
equilibrium between free and bound compound is reached after one hour. However, it cannot be
ruled out that precipitation of the compound also contributes to the change in fluorescence meaning
that an amount of 15% DMSO is insufficient to hold compound 52 permanently in solution.
In general, the estimation of %PPB by fluorescence measurements is impossible in case of
compounds 51 and 52, mainly due to the lack of emission at physiological pH in combination with
precipitation, effects of DMSO on the structure and solubility of BSA.
82
Chapter 4
4.3.2.4 Determination of lipophilicity parameters by HPLC
As all the aforementioned approaches for the estimation of protein binding failed, HPLC-based
methods (experiments performed in cooperation with Origenis GmbH, Martinsried, Germany) were
used to determine lipophilicity parameters for the prediction of PPB based on the work of Valko et.
al. [Valko et al., 1997; Valko et al., 2000; Valko et al., 2003; Valko et al., 2011]. As displayed in
Table 4.3 compounds 31, 32 and 51 (Figure 4.13) were estimated to bind to human serum albumin to
almost 100%.
Figure 4.13: Chemical structures of ABCG2 modulators 31, 32 and 51 with a molecular weight of 640, 772 and 666 Da,
respectively.
Table 4.3: HSA binding affinities and chromatographically determined lipophilicity parameters of
compounds 31, 32 and 51, respectively, each calculated in 2 independent experiments.
Cpd.
31
32
51
a
logK
logkA
Confid.b
%PPBa
(HSA)
(HSA)
HSA
CHIc
98.37
1.78
4.96
1
123.61
99.02
2.00
5.19
0
21.45
-0.56
2.62
1
98.07
1.71
4.89
1
98.58
1.84
5.02
2
98.33
1.77
4.95
1
105.03
125.35
Confid.b
CHI
Confid.b
logd
logd
IAMd
IAM
3.59
2
74.50
1
74.61
1
57.19
1
58.26
1
62.40
2
62.57
1
3.17
3.62
0
2
Calculated percental plasma protein binding; HPLC conditions: solvent A (5% (v/v) isopropanol in 50 mM ammonium
acetate buffer; pH 7,4)/isopropanol (100%): 0 min: 100/0, 3.5 min: 0/30, 4.5 min: 0/30, 4.6 min: 100/0, 6 min: 100/0.
b
c
Confidence: 2 = without any doubt, 1 = likely, 0 = unlikely. A C18 column (40 °C) was used with an ammonium acetate
buffer [50 mM; pH 7.4] + 5% (v/v) MeCN as solvent (solvent B) under the following gradient conditions: solvent B/MeCN:
d
0 min: 100/0, 3 min: 0/100, 4 min: 0/100, 4.2 min: 100/0, 6 min: 100/0. IAM column at 40 °C with the following gradient:
solvent B/MeCN: 0 min: 100/0, 5 min: 0/100, 5.5 min: 0/100, 6.5 min: 100/0, 8.2 min: 100/0.
83
The HSA binding values were derived from the gradient retention times and converted in its linear
free energy-related logK value using data from a calibration set of molecules:
log
(e ua on 5)
Percental value data and logK values can be converted to binding affinity constant (logkA) according
to the following formula:
log
(e ua on 6
kA represents the binding affinity to HSA under the assumption that binding occurs exclusively to
HSA, a binary complex is formed between the ligand and HSA, and the albumin concentration in
plasma is around 0.0006 M ([HSA]) [Kratochwil et al., 2002; Borchardt et al., 2004].
Characterizing interactions of drugs via immobilized artificial membrane (IAM) chromatography is an
efficient and time-saving method to obtain membrane interaction parameters of drugs, enabling the
estimation of a compound’s lipophilicity. With a calibration set of standard compounds, the gradient
retention times can be converted to chromatographic hydrophobicity index (CHI) values (CHI IAM).
As CHI correlates closely with traditional octanol-buffer distribution coefficient logD, CHI values were
projected to a logarithmic scale by an intralaboratory equation that is more appropriate for a
comparison (logd).
In general, the investigated ABCG2 modulators show very high protein (>98%) and membrane
binding due to high lipophilicity.
4.3.3
pH-dependent fluorescence of ABCG2 modulators
As described in 4.3.2.3, selected ABCG2 modulators of the triazole series like compound 52 show an
increase in fluorescence with decreasing pH values of the solvent.
A strong acid milieu of pH 1.5 even results in fluorescence emission in case of the biphenyl
compound 12, which is non-fluorescent in neutral milieu (Figure 4.14).
This fact suggests that the pH-dependent increase in fluorescence results from the protonation at the
common quinoline substructure, thereby extending the conjugation of the π-electron system.
84
Chapter 4
700
A
600
600
500
500
realtive intensity
relative intensity
700
400
300
200
400
300
200
100
100
0
0
400
450
500
550
wavelength [nm]
600
B
400
450
500
550
600
wavelength [nm]
Figure 4.14: Fluorescence spectra of compounds 12 [1 µM] (A) and 52 [1 µM] (B) in DMSO (black) and DMSO/TFA (aq.) at
pH 1.5 (red) after excitation at 370 nm; Ex/Em slit 10. In case of 52, a slight shift of the emission maximum to the blue was
visible with the addition of TFA.
ACD labs chemistry software identified the quinoline as a weak base with a pKa value around 1,
whereas the 1,2,3-triazole substructure is not involved in an pH dependent fluorescence according to
these calculations at pH 1.5 (Figure 4.15).
Figure 4.15: Calculated pKa values for compounds 12 and 52 by ACD/Labs chemistry software 12.0.
However, these pKa values have to be examined critically as they can strongly depend on the
electronic charge distribution of the compound resulting in enormous differences in acidity between
the ground and excited states [Marciniak et al., 1992] as is the case for six-membered aromatic
nitrogen heterocyclic compounds, which are generally known to be more basic in the lowest excited
state than in the ground state usually by at least five orders of magnitude [Schulman and
Capomacchia, 1973].
85
Intramolecular proton transfer of organic compounds can result in drastic changes in their
fluorescence properties. Detailed reaction dynamics and the influence of the microenvironment on
the mechanism of reaction are often not completely understood.
The
polymer
photostabilizer
Tinuvin
P
(2-(2’-hydroxy-5’-methylphenyl)benzotriazole),
the
photostability of which has been attributed to a very fast reversible intramolecular proton transfer, is
shown in Figure 4.16. Keto-enol-like tautomers can be distinguished and different fluorescence
properties (in various solvents) are due to different energy states: ground and excited states
[Wiechmann et al., 1991].
Figure 4.16: Keto- and enol-like tautomers of
Tinuvin P.
Related results were obtained for 2-(2‘-hydroxyphenyl)oxazole and 2-(2’-hydroxyphenyl)-4methylthiazole [LeGourrierec et al., 1998].
A similiar pH- dependent tautomeric mechanism as well as solvent-dependent influences on the
energy state in terms of fluorescence properties are also conceivable for the synthesized ABCG2
modulators both with the quinoline and the triazole substructre (Figure 4.17).
Figure 4.17: Hypothetical highly protonated form of compound 51.
86
4.4
Chapter 4
The attempts to determine protein binding failed irrespective of the applied methods (equilibrium
dialysis, ITC, ultracentrifugation, fluorescence spectroscopy). Very poor water solubility of the
investigated ABCG2 modulators is the major reason for insufficient drug-like properties. High
molecular weights (between 600 and 900 Da) and high logP values usually result in poor absorption
and permeation. In addition the extent of plasma protein binding is probably very high, as predicted
on the basis of HPLC measurements. However, despite these unfavorable properties, the compounds
show high potency in cell-based in vitro assays suggesting that they are either actually much more
potent than calculated on the basis of the added concentration, i.e., the effective concentrations of
free compounds are much lower, or the binding to proteins is beneficial to provide a reservoir of the
compounds in the aqueous systems. It has to be further elucidated whether these ABCG2 modulators
are suitable for in vivo studies in nude mice.
To come to more ‘drug-like’ compounds, the chemical structures have to be modified by introducing
more hydrophilic moieties and/or by reducing the molecular weight.
4.5
87
References
Abbott, N. J., Patabendige, A. A., Dolman, D. E., et al. Structure and function of the blood-brain
barrier. Neurobiol. Dis. 2010, 37(1), 13-25.
Anderson, N. L. and Anderson, N. G. The human plasma proteome: history, character, and diagnostic
prospects. Mol. Cell. Proteomics 2002, 1(11), 845-867.
Borchardt, R. T., Kerns, E., Lipinski, C. A., et al. Pharmaceutical Profiling in Drug Discovery for Lead
Selection. American Association of Pharmaceutical Scientists, Arlington, VA, 2004.
Bujacz, A. Structures of bovine, equine and leporine serum albumin. Acta Crystallogr. D Biol.
Crystallogr. 2012, 68(Pt 10), 1278-1289.
Estey, T., Kang, J., Schwendeman, S. P., et al. BSA degradation under acidic conditions: a model for
protein instability during release from PLGA delivery systems. J. Pharm. Sci. 2006, 95(7),
1626-1639.
Galecki, K., Despotovic, B., Galloway, C., et al. Binding of harmane to human and bovine serum
albumin: fluorescence and phosphorescence study. Biotechnol. Food Sci. 2012 2012, 76(1), 312.
Ge, S., Kojio, K., Takahara, A., et al. Bovine serum albumin adsorption onto immobilized
organotrichlorosilane surface: influence of the phase separation on protein adsorption
patterns. J. Biomater. Sci. Polym. Ed. 1998, 9(2), 131-150.
Kratochwil, N. A., Huber, W., Muller, F., et al. Predicting plasma protein binding of drugs: a new
approach. Biochem. Pharmacol. 2002, 64(9), 1355-1374.
Ladokhin, A. S. Fluorescence Spectroscopy in Peptide and Protein Analysis. Encyclopedia of Analytical
Chemistry, John Wiley & Sons Ltd. 2000, 5762 – 5779.
LeGourrierec, D., Kharlanov, V., Brown, R. G., et al. Excited-state intramolecular proton transfer
(ESIPT) in 2-(2 '-hydroxyphenyl)pyridine and some carbon-bridged derivatives. J Photoch
Photobio A 1998, 117(3), 209-216.
Lipinski, C. A. Drug-like properties and the causes of poor solubility and poor permeability. J.
Pharmacol. Toxicol. Methods 2000, 44(1), 235-249.
Lipinski, C. A., Lombardo, F., Dominy, B. W., et al. Experimental and computational approaches to
estimate solubility and permeability in drug discovery and development settings. Adv. Drug
Deliv. Rev. 2001, 46(1-3), 3-26.
Lund, L., Berlin, A. and Lunde, K. M. Plasma protein binding of diphenylhydantoin in patients with
epilepsy. Agreement between the unbound fraction in plasma and the concentration in the
cerebrospinal fluid. Clin. Pharmacol. Ther. 1972, 13(2), 196-200.
Majorek, K. A., Porebski, P. J., Dayal, A., et al. Structural and immunologic characterization of bovine,
horse, and rabbit serum albumins. Mol. Immunol. 2012, 52(3-4), 174-182.
Marciniak, B., Kozubek, H. and Paszyc, S. Estimation of Pka* in the 1st Excited Singlet-State - a
Physical-Chemistry Experiment That Explores Acid-Base Properties in the Excited-State. J.
Chem. Educ. 1992, 69(3), 247-249.
88
Chapter 4
Miners, J. O., Foenander, T., Wanwimolruk, S., et al. The effect of sulphinpyrazone on oxidative drug
metabolism in man: inhibition of tolbutamide elimination. Eur. J. Clin. Pharmacol. 1982,
22(4), 321-326.
Moriyama, Y., Ohta, D., Hachiya, K., et al. Fluorescence behavior of tryptophan residues of bovine
and human serum albumins in ionic surfactant solutions: a comparative study of the two and
one tryptophan(s) of bovine and human albumins. J. Protein Chem. 1996, 15(3), 265-272.
Mungall, D., Wong, Y. Y., Talbert, R. L., et al. Plasma protein binding of warfarin: methodological
considerations. J. Pharm. Sci. 1984, 73(7), 1000-1001.
Pabbathi, A., Patra, S. and Samanta, A. Structural transformation of bovine serum albumin induced
by dimethyl sulfoxide and probed by fluorescence correlation spectroscopy and additional
methods. Chemphyschem 2013, 14(11), 2441-2449.
Pajouhesh, H. and Lenz, G. R. Medicinal chemical properties of successful central nervous system
drugs. NeuroRx. 2005, 2(4), 541-553.
Patra, S., Santhosh, K., Pabbathi, A., et al. Diffusion of organic dyes in bovine serum albumin solution
studied by fluorescence correlation spectroscopy. Rsc Adv. 2012, 2(14), 6079-6086.
Peters, T. Serum-Albumin. Adv. Protein Chem. 1985, 37, 161-245.
Peterson, G. M., McLean, S., Aldous, S., et al. Plasma protein binding of phenytoin in 100 epileptic
patients. Br. J. Clin. Pharmacol. 1982, 14(2), 298-300.
Pierce, M. M., Raman, C. S. and Nall, B. T. Isothermal titration calorimetry of protein-protein
interactions. Methods 1999, 19(2), 213-221.
Ptachcinski, R. J., Venkataramanan, R. and Burckart, G. J. Clinical pharmacokinetics of cyclosporin.
Clin. Pharmacokinet. 1986, 11(2), 107-132.
Reichel, A. The role of blood-brain barrier studies in the pharmaceutical industry. Curr. Drug Metab.
2006, 7(2), 183-203.
Rolan, P. E. Plasma protein binding displacement interactions-why are they still regarded as clinically
important? Br. J. Clin. Pharmacol. 1994, 37(2), 125-128.
Santos, H. A., Manzanares, J. A., Murtomaki, L., et al. Thermodynamic analysis of binding between
drugs and glycosaminoglycans by isothermal titration calorimetry and fluorescence
spectroscopy. Eur. J. Pharm. Sci. 2007, 32(2), 105-114.
Schulman, S. G. and Capomacchia, A. C. Dual Fluorescence of Quinolinium Cation. J. Am. Chem. Soc.
1973, 95(9), 2763-2766.
Stegemann, S., Leveiller, F., Franchi, D., et al. When poor solubility becomes an issue: From early
stage to proof of concept. Eur. J. Pharm. Sci. 2007, 31(5), 249-261.
Takehara, K., Yuki, K., Shirasawa, M., et al. Binding properties of hydrophobic molecules to human
serum albumin studied by fluorescence titration. Anal. Sci. 2009, 25(1), 115-120.
Valko, K., Bevan, C. and Reynolds, D. Chromatographic Hydrophobicity Index by Fast-Gradient RPHPLC: A High-Throughput Alternative to log P/log D. Anal. Chem. 1997, 69(11), 2022-2029.
89
Valko, K., Du, C. M., Bevan, C. D., et al. Rapid-gradient HPLC method for measuring drug interactions
with immobilized artificial membrane: comparison with other lipophilicity measures. J.
Pharm. Sci. 2000, 89(8), 1085-1096.
Valko, K., Nunhuck, S., Bevan, C., et al. Fast gradient HPLC method to determine compounds binding
to human serum albumin. Relationships with octanol/water and immobilized artificial
membrane lipophilicity. J. Pharm. Sci. 2003, 92(11), 2236-2248.
Valko, K. L., Nunhuck, S. B. and Hill, A. P. Estimating unbound volume of distribution and tissue
binding by in vitro HPLC-based human serum albumin and immobilised artificial membranebinding measurements. J. Pharm. Sci. 2011, 100(3), 849-862.
Veber, D. F., Johnson, S. R., Cheng, H. Y., et al. Molecular properties that influence the oral
bioavailability of drug candidates. J. Med. Chem. 2002, 45(12), 2615-2623.
Venkataramanan, R., Swaminathan, A., Prasad, T., et al. Clinical pharmacokinetics of tacrolimus. Clin.
Pharmacokinet. 1995, 29(6), 404-430.
Vivian, J. T. and Callis, P. R. Mechanisms of tryptophan fluorescence shifts in proteins. Biophys. J.
2001, 80(5), 2093-2109.
Waterhouse, R. N. Determination of lipophilicity and its use as a predictor of blood-brain barrier
penetration of molecular imaging agents. Mol. Imaging Biol. 2003, 5(6), 376-389.
Wiechmann, M., Port, H., Frey, W., et al. Time-Resolved Spectroscopy on Ultrafast Proton-Transfer in
2-(2'-Hydroxy-5'-Methylphenyl)Benzotriazole in Liquid and Polymer Environments. J Phys
Chem-Us 1991, 95(5), 1918-1923.
90
Chapter 4
Chapter 5
5
Characterization of human brain
tumor cell lines
5.1
Malignant brain tumors
5.1.1
Classification
According to the classification of central nervous system (CNS) malignancies recommended by the
World Health Organization (WHO), brain tumors are grouped in tumor types and grades ranging from
grade I (least malignant) to grade IV (most malignant) based on their histopathology [Louis et al.,
2007]. Brain cancer can be categorized into primary brain tumors, originating from a brain cell type,
and secondary tumors, i.e. brain metastases originating from malignancies in the periphery. 30% of
all primary brain tumors are gliomas, descending from the glial (supportive) tissue [Dolecek et al.,
2012]. Gliomas are grouped in several subtypes, including astrocytomas, ependymomas and
oligodendrogliomas.
Glioblastoma multiforme (GBM) is the most common and aggressive form of all malignant primary
brain tumors with a median survival time of only 15 months [Wen and Kesari, 2008]. It belongs to the
group of astrocytic tumors and is classified as grade IV. About 15% of all brain tumors are GBM.
The aggressive and malignant medulloblastomas belong to the group of embryonal tumors and
commonly occur in the cerebellum. Medulloblastomas and gliomas are the most common brain
tumors in children (20% of childhood brain tumors) and carry the most malignant WHO designation
(grade IV) [Huse and Holland, 2010; Cook and Freedman, 2012].
5.1.2
Incidence and mortality
As results of a project coordinated by the International Agency for Research on Cancer (IARC; Lyon,
France) [Ferlay et al., 2013] the incidence and mortality data of different cancers in 40 European
countries were presented for the year 2012. There were 3.45 million new cases of cancer and an
92
Chapter 5
estimated number of 1.75 million deaths from cancer in Europe in 2012 with female breast,
colorectal, prostate and lung cancer as the most common manifestations. The estimated number of
new brain cancer cases in Europe was 57,100 with a concomitant mortality of 45,000.
Though new brain tumor incidences represent only 2% of all estimated new cancer cases, efficient
treatment is urgently required, as the trend of new cases is rising (190,000 sufferers in the year 2002
worldwide, 238,000 in 2008) [Ferlay et al., 2010] and the five-year relative survival rates (percentage
of people who are alive 5 years after their diagnosis) are still very low. Worldwide, every day about
650 people are diagnosed with a malignant brain tumor.
5.1.3
The blood-brain barrier and brain cancer therapy
The blood-brain barrier separates the brain and cerebrospinal fluid from the blood and is composed
of high-density endothelial cells. As a regulatory barrier, it controls the entry and elimination of ions,
neurotransmitters, macromolecules, neurotoxins or water soluble nutrients from and into the brain
and, hence, sustains ionic homeostasis [Abbott et al., 2010]. Tight junctions are cell-cell junctional
complexes in the apical region of cell membrane and form the intercellular barrier between the
epithelial cells, thereby, almost completely obstructing paracellular diffusion under normal
conditions [Stamatovic et al., 2008]. Transcellular pathways over the BBB include passive diffusion,
adsorptive-mediated transcytosis and the active transport of molecules via various transportermediated shuttle systems (Figure 5.1).
transcellular
pathways
water,
lipophilic
agents
paracellular
transport
water,
glycerol,
urea
receptormediated
transcytosis
insulin,
transferrin
carriermediated
transcytosis
glucose,
amino acids
efflux
pumps
vinca
alkaloids,
ciclosporin A
adsorptive
transcytosis
plasma
proteins,
kationic
compounds
+
Blood
+
+
Endothelial cells
Brain
Figure 5.1: Schematic illustration of transport ways and mechanisms at the blood-brain barrier. Tight junctions are
displayed in red.
Characterization of human brain tumor cell lines
93
The transporters expressed in the BBB comprise carrier-mediated, receptor-mediated and active
efflux transporters [Pardridge, 2005] including organic anion and cation transporter systems [Sekine
et al., 2000], nucleoside transporters and ABC efflux transporter like P-gp, BCRP and MRPs [Lee et al.,
2001]. The BBB protects the CNS from xenobiotics including a large variety of pharmacotherapeutics.
As a consequence, the medical treatment of CNS diseases like brain tumors is compromised. Several
strategies were developed to overcome or bypass the BBB, aiming at improved therapies of CNS
malignancies.
BCNU (1,3-bis(2-chloroethyl)-1-nitrosourea; carmustine), an alkylating agent of the group of
nitrosoureas included in standard regimen for nearly 30 years [Walker and Hurwitz, 1970], as well as
the DNA methylating agent temozolomide (Temodal®) are the drugs frequently used in systemic
chemotherapy [Newlands et al., 1997; Stupp et al., 2005]. Promising strategies for an enhanced drug
delivery into the brain have been developed, for instance, the therapeutic application of ‘vectors’
such as peptides [Rousselle et al., 2001] and viruses [Pertschuk et al., 2012; Robbins et al., 2012] or
nanoparticles as carriers for the transport of packaged drugs to the brain [Martin-Banderas et al.,
2011; Orthmann et al., 2012; Tosi et al., 2013].
Another field of current research in brain tumor treatment is the use of monoclonal antibodies. Since
May 2009, bevacizumab (Avastin®) is approved by the Food and Drug Administration (FDA) for
second-line treatment of glioblastoma in the USA [Cohen et al., 2009]. It is a recombinant humanized
monoclonal antibody that binds to human vascular endothelial growth factor (VEGF) inhibiting
intratumoral angiogenesis and, thereby, tumor growth and proliferation [Shih and Lindley, 2006].
Advances in chemotherapy also include more invasive approaches namely intracerebral (i.c.) and
intracerebroventricular (i.c.v.) applications as well as the direct placement of the anticancer agent
into the tumor cavity via convection enhanced delivery (CED) by means of a catheter or an implanted
drug reservoir [Saito and Tominaga, 2012]. Biodegradable implants have been approved for the
treatment of CNS tumors. The Gliadel® wafer, containing BCNU embedded in a polyanhydride matrix,
is indicated in newly-diagnosed high-grade glioma and recurrent GBM patients. It is planted into the
surgical cavity post surgery to release the cytostatic directly to the tumor site [Panigrahi et al., 2011].
In general, invasive methods mostly require a complex surgical intervention and are precarious not
least because of unpredictable risks of infection, injury and long-term effects.
A better understanding of the pharmacology of brain tumors and the growth behavior and pathology
of available human brain tumor cell lines in vitro is of major importance with respect to optimization
of treatment therapies aiming at extending the average survival times and reducing the size and
progression of malignant brain tumors. During this work, three different brain tumor cell lines were
comprehensively characterized in vitro and partially in vivo.
94
Chapter 5
5.2
5.2.1
Drugs and chemicals
PBS (phosphate buffered saline) was made of 8.0 g/L NaCl, 1.0 g/L Na2HPO4 · 2 H2O, 0.20 g/L KCl,
0.20 g/L KH2PO4 and 0.15 g/L NaH2PO4 · H2O. The pH-value was adjusted to 7.4.
Bouin’s solution (fixative for histopathology) was made of saturated picric acid, formaldehyde and
glacial acetic acid in a ratio of 15:5:1. Giemsa solution for chromosomal staining was from AppliChem
(Darmstadt, Germany) and Tris base as buffer constituent from usb (Cleveland, OH). The
PerfectBlue™ Double gel system Twin S for SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel
electrophoresis) as well as the nitrocellulose membrane and the PerfectBlue™ ‘Semi-Dry’ electro blot
apparatus for Western Blotting were obtained from Peqlab (Erlangen, Germany). Topotecan,
etoposide, doxorubicin, methotrexate and vinblastine were purchased from Sigma (Munich,
Germany). Paclitaxel (6 mg/mL; Bristol-Meyers Squibb, Wien, Austria) was obtained from the
pharmacy of the University Hospital Regensburg. A solution of mitoxantrone (1 mM) was obtained by
diluting Novantron® (Wyeth Pharma, Münster, Germany) in 70% ethanol.
If not otherwise stated, chemicals (p.a. quality) were obtained from Merck (Darmstadt, Germany)
and antibodies from Santa Cruz Biotechnology (Heidelberg, Germany). Milli-Q system (Millipore,
Eschborn, Germany) was used for the purification of water.
5.2.2
Cell lines and culture conditions
The human brain tumor cell lines SW 1783 (ATCC® HTB-13™; astrocytoma, grade III), Daoy (ATCC®
HTB-186™; desmoplastic cerebellar medulloblastoma) and LN-18 (ATCC® CRL-2610™; glioblastoma
grade IV) as well as the murine cell line P388D1 (ATCC® CCL-46™; murine lymphoid neoplasm with
macrophage-like morphology) were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Sigma,
Munich, Germany) supplemented with 10% FCS (fetal calf serum; Biochrom, Berlin, Germany), 3.7 g/L
sodium hydrogen carbonate and 110 mg/L sodium pyruvate (Serva, Heidelberg, Germany).
Immortalized human microvascular endothelial cells (HMEC-1) [Ades et al., 1992] were kindly
provided by Prof. Dr. J. Heilmann (Institute of Pharmaceutical Biology, University of Regensburg) and
cultured in DMEM containing 15% FCS and 2 mM L-glutamine (Biochrom).
Cells were cultured in a water-saturated atmosphere (95% air, 5% CO2) at 37 °C in 25-cm2 or 75-cm2
cell culture flasks purchased from Nunc (Wiesbaden, Germany), Greiner (Frickenhausen, Germany)
and Sarstedt (Nümbrecht, Germany), respectively.
5.2.3
95
Cell staining
Microscopic slides were immersed in 200 mL 70% ethanol, containing 6 mL of concentrated HCl, for
24 h, washed with distilled water and dried. Dry coverslips were placed between filter papers and
sterilized by autoclaving (121 °C, 15 psi) [Rooney and Czepulkowski, 1992].
Cells were seeded on the prepared slides for 1 day (5% CO2, 37 °C), afterwards fixed for 1 h with
Carnoy´s solution (ethanol/chloroform/glacial acetic acid 6:3:1) and stored in PBS containing
0.1% NaN3. Sections were evaluated with a BH-2 microscope (Olympus, Hamburg, Germany) after
Papanicolaou staining [Takahashi, 1987] using ScopePhoto 3.0 software.
5.2.4
Genetic stability - Karyology
The assay was performed as described previously [Jarzyna, 2007]: Cells were cultured on immersed
and sterilized microscopic slides in Quadriperm 4-well lux-multiplates (Greiner) up to a confluency of
15-25% and afterwards overlaid with cell culture medium containing 0.5 µg/mL colcemide to arrest
mitosis of the cells during metaphase.
After an incubation time of 3 h (37 °C, 5% CO2), the medium was removed and a hypotonic potassium
chloride solution (75 mM) preheated to 37 °C was carefully added to lyse the cells. After 30 min, an
equal volume of ice-cold fixative (methanol/glacial acetic acid 3:1) was added and carefully removed
at once. After two additional fixation steps on ice (last step: 10 min fixation) the fixative was
removed and the slides were dried. Slides were stained for 8 min with a diluted Giemsa staining
solution (10 mL filtered Giemsa solution, 90 mL of 25 mM KH2PO4, pH 6.8), thoroughly washed, dried
and mounted with DePex (EMS, Hatfield, PA). Chromosome numbers of at least 50 cells in metaphase
were determined per cell line with an Olympus BH-2 microscope using 40x and 100x Planapo oil
immersion objectives (Zeiss, Jena, Germany).
5.2.5
Chemosensitivity against common cytostatic drugs and
growth kinetics
The aforementioned human brain tumor cell lines were investigated for their chemosensitivity
against common cytostatic drugs by means of a crystal violet assay as described in section 3.3.6.
In vitro growth behavior was determined in a similar manner: Cells were seeded in 96-well plates,
fixed with glutardialdehyde after different times of incubation (37 °C, 5% CO2) and stained with
crystal violet. The absorbance, proportional to cell mass/number, was measured at 580 nm using a
GENios Pro microplate reader (TECAN Deutschland GmbH, Crailsheim, Germany) and plotted on a
96
Chapter 5
log-linear scale against the incubation time. The population doubling time was determined along the
exponential phase of the curve.
5.2.6
Tumorigenicity and growth kinetics of subcutaneous tumors
in nude mice
For tumor cell implantation into nude mice, human brain tumor cells were washed twice with 5 mL of
pre-warmed serum-free culture medium and carefully harvested using a cell scratcher. After
centrifugation at 1,000 rpm for 7 min at room temperature, using a Minifuge 1.0R (Heraeus
Instruments GmbH, Hanau, Germany), the obtained cell pellet was resuspended in serum-free
medium and the cell number was adjusted to a density of 1-3 x 106 cells per 100 µL by means of a
Casy® 1 TTC cell counter (Schärfe System GmbH, Reutlingen, Germany) or a Neubauer counting
chamber.
100 µL of the cell suspension were injected under the thoracic dermis of each mouse. Once a week,
the bodyweight was controlled and the tumor diameters were monitored with a sliding caliper. To
ensure tumor growth over various passages, 1 mm3 tumor pieces were prepared from an excised
non-necrotic tumor and again transplanted into nude mice by means of a trocar.
5.2.7
Histology
Solid tumors were prepared for histological staining by the following procedure:
After dissection, tumor tissue was fixed in Bouin´s solution and embedded in paraffin after
dehydration with increasing concentrations of ethanol by a standard procedure (described in detail in
[Müller, 2007]). Sections (6 μm) were cut by a Leica RM2255 microtome (Leica, Bensheim, Germany)
and mounted on microscope slides. After deparaffinization with xylene and rehydration in decreasing
concentrations of ethanol, slides were subsequently stained by the procedure of Masson-Goldner
(MG) or with haematoxylin-eosin (HE) [Mulisch and Welch, 2010], dried and mounted with DePeX.
5.2.8
Induction of BCRP overexpression in brain tumor cells
To determine the ABCG2 transporter expression in the aforementioned human brain tumor cells
Western Blot (WB) analysis of total cell lysates was performed. BCRP-positive cell types were further
cultured and treated with topotecan at increasing concentrations over a period of approximately
10 weeks to amplify the transporter expression.
5.2.9
97
ABC transporter detection by Western Blot analysis
5.2.9.1 Cell lysis and protein quantification
Cell lines were lysed by physical disruption. Cells were harvested with a cell scratcher and washed
three times with PBS. After centrifugation, the cell pellet was homogenized in PBS containing a
protease inhibitor cocktail (SIGMAFAST™ Tablets; Sigma, Munich, Germany) followed by 5-6
repetitive freeze-thaw cycles in liquid nitrogen to disrupt the cells through ice crystal formation.
After homogenization the protein content of each cell lysate was determined in a Bio-Rad Protein
Assay (Bio-Rad Laboratories, Munich, Germany) based on the method of Bradford [Bradford, 1976]
according to the standard procedure described in the manual. UV detection was performed with a
Cary 100 UV-Vis spectrophotometer (Varian, Darmstadt, Germany).
5.2.9.2 SDS-PAGE and Western blot
Proteins were analyzed by SDS-PAGE. The basic procedure was that of Laemmli [Laemmli, 1970],
using a 12% acrylamide resolving gel. The composition of all used buffers (in millipore water) was as
follows:
Buffer A:
1.5 M Tris, 0.4% (m/v) SDS, pH 8.8
Buffer B:
0.5 M Tris, 0.4% (m/v) SDS, pH 6.8
Running buffer:
0.025 M Tris, 0.2 M glycine, 0.1% (m/v) SDS, pH 8.3
Blotting buffer:
(always freshly prepared) 0.025 M Tris, 0.2 M glycine, 20% (v/v) methanol
TBS-T buffer:
0.02 M Tris, 0.14 M NaCl in millipore water (pH 7.6), 0.1% (v/v) Tween 20
(Carl Roth GmbH, Karlsruhe, Germany)
Sample buffer (6x):
0.15 M Tris, 1.0 M glycine, 0.6% (m/v) SDS, 9.0% (v/v) glycerol, 0.1% (m/v)
bromophenol blue, 2% (v/v) β-mercaptoethanol, pH 6.8
Leammli buffer (6x):
0.38 M Tris, 12% (m/v) SDS, 60% (v/v) glycerol, 0.06% (m/v) bromophenol
blue, 10 mM DTT
A 12% separation gel mixture for one gel contained 2.65 mL of millipore water, 2 mL of buffer A and
3.1 mL of acrylamide/bisacrylamide 30% solution (acrylamide/bisacrylamide 29/1; Sigma, Munich,
Germany). Polymerization was started by addition of 3.5 μL of N,N,N',N'-tetramethylethylenediamine
(TEMED; Serva, Heidelberg, Germany) and 35 μL of a 10% (m/v) ammonium peroxodisulfate solution
(APS; Serva).
The mixture was filled into gel chambers (10 x 10 x 0.8 cm) and covered with a layer of
water-saturated isobutyl alcohol. After complete polymerization (ca. 45 min) and discarding the
98
Chapter 5
isobutyl alcohol layer, a stacking gel (3%) was poured on top of the separation gel. The stacking gel
mixture contained 3.25 mL of millipore water, 1.25 mL of buffer B and 0.5 mL of
acrylamide/bisacrylamide 30% solution. Polymerization was initiated as described above.
A defined amount of protein per cell lysate was mixed with 1/6 volume of Laemmli buffer (6x) or
sample buffer (6x) and heated for 5 min at 100 °C before filling in the stacking gel pockets.
Electrophoresis was performed in a PerfectBlue™ gel electrophoresis chambers filled with running
buffer at 150 V for approximately 90 min.
The unstained protein marker peqGOLD Protein Marker I (14.4-116 kDa; Peqlab, Erlangen, Germany)
or the Biotinylated Protein Ladder (9-200 kDa) with the Anti-biotin HRP-linked Antibody (1 : 1,000)
from Cell Signaling Technology (Danvers, MA) were used.
WB was performed in a Perfect-Blue™ ‘Semi-Dry’ electro blot apparatus by placing the gels on top of
a nitrocellulose membrane (0.2 μm) between 6 filter slides soaked in blotting buffer. The proteins
were transferred to the membrane by blotting for 45 min at 250 mA. Then, the membrane was
blocked by washing in 5% fat-free milk (in TBS-T buffer; block solution) for 2 h at room temperature
or overnight at 4 °C. After three washing steps (5 min each) in TBS-T buffer, the primary antibody was
incubated with the membrane for 2 h (rt) in block solution. After washing, the secondary antibody,
conjugated with horseradish peroxidase, was diluted in block solution, added to the membrane and
shaken at rt for 1 h. Immunoreactive bands were visualized via enhanced chemoluminescence (ECL)
using the Pierce® ECL Western Blotting substrate kit (Thermo Scientific, Dreieich, Germany) according
to the manufacturer’s instructions.
Blots and gels were scanned using a GS-710 imaging densitometer (Bio-Rad Laboratories, Munich,
Germany). Quantity One (Bio-Rad) software was used for band analysis.
A total protein content of 20 µg was blotted for each cell line. The primary rat monoclonal
antibody ABCG2 (BXP-53) was used in a 1: 1,000 dilution, the secondary AB (goat anti-rat IgG-HRP) in
a 1: 10,000 dilution.
5.2.10 ABC transporter detection by flow cytometry
To further characterize the ABC transporter expression, a flow cytometric FACS (fluorescence
activated cell sorting) assay was performed. Confluent cells were trypsinized into a 1.5 mL Eppendorf
reaction vessel, suspended in PBS and centrifuged at 1,000 g for 5 min. After resuspension in PBS
supplemented with 10% FCS, cells were counted in a Casy® 1 TTC cell counter and the density was
adjusted to a number of 2 x 106 cell/mL with PBS/FCS. Aliquots of 500 µL were taken and incubated
with the primary antibody in an appropriate dilution. After 1 h of incubation at 4 °C under shaking,
99
three washing steps with PBS (1,000 g, 5 min each) followed. Cells were resuspended in a solution of
3% (m/v) BSA (bovine serum albumin) in PBS containing the phycoerythrin (PE)-conjugated
secondary antibody and incubated for an additional hour in the dark. Then, cells were washed three
times with PBS, resuspended in 500 µL BSA/PBS and measured at a FacsCalibur™ (Becton Dickinson,
Heidelberg, Germany). Instrument settings were as follows:
FSC: E-1; SSC: 350, Fl-2, high flow. WinMDI 2.9 software was used for data analysis.
Antibodies and dilutions used for FACS analysis:
ABCB1:
Primary AB:
Mdr (C-19) goat polyclonal antibody, 1:50
Secondary AB: rabbit anti-goat IgG-PE, 1:100
ABCC1:
Primary AB:
MRP1 rabbit polyclonal antibody (abbiotec; Hoelzel Diagnostika
GmbH, Köln, Germany), 1:30
Secondary AB: goat anti-rabbit IgG-PE (Rockland Inc.; Biomol GmbH, Hamburg,
Germany), 1:100
ABCG2:
Primary AB:
ABCG2 (BXP-53) rat monoclonal antibody, 1:50
Secondary AB: goat anti-rat IgG-PE, 1:100
100
Chapter 5
5.3
5.3.1
Morphology
Doay, LN-18 and SW 1783 cells, grown on tissue slides, were stained according to the method of
Papanicolaou to illustrate plasma turquoise and nuclei blue (Figures 5.2-5.4).
A
B
Figure 5.2: Microscopic images (Papanicolaou staining) of human Daoy cells (passage 7) with a high mitotic activity (A,
objective: 20x); B) Magnification (objective: 40x) of highly malignant multi-nucleated cells.
A
B
Figure 5.3: Microscopic images (Papanicolaou staining) of human LN-18 cells (passage 18); the cells show distinctive
anisomorphism, an increased nucleoplasmatic ratio (A, objective: 20x) and multiple prominent nucleoli (B, lower left
corner; objective: 40x).
A
101
B
Figure 5.4: Microscopic images (Papanicolaou staining) of human SW 1783 cells (passage 42); highly anisomorphic cells (A,
objective: 20x) with a high mitotic activity and multiple prominent nucleoli (B, objective: 40x).
5.3.2
In vitro growth of brain tumor cells
To evaluate growth characteristics like the doubling time or saturation density of the aforementioned
brain tumor cell lines, growth curves were generated before starting any experiments. The growth of
cells in culture is characterized by a lag phase, the time it takes for the cells to recover from
subculture, attach and spread, a log phase in which an exponential increase in cell density arises and
a stationary phase in which cell proliferation slows down due to the cell culture becoming confluent.
The growth of Daoy (5th passage), LN-18 (7th passage) and SW 1783 (31st passage) cells over two
weeks is shown in Figure 5.5.
4
absorbance A580
2
1
Figure 5.5: In vitro growth curves of LN-18 (squares), Daoy (triangles)
and SW 1783 (circles) cells determined spectrophotometrically (UV
detection, 580 nm).
0.1
0
100
200
300
400
hours after seeding
The Daoy and LN-18 cell lines show similar growth rates with a mean doubling time of 40 h and 47 h,
respectively. The SW1783 cell line shows the slowest proliferation kinetics with a mean doubling time
of 64 h.
102
Chapter 5
Doubling times of the three brain tumor cell lines did not significantly vary over different in vitro
passages (data not shown) indicating that growth rates and chemosensitivities are independent of
the utilized passage.
5.3.3
Aneuploidy
A prominent feature of cancer cells is an abnormal number of chromosomes (aneuploidy) caused by
genomic instability inherent to most cancers. It occurs through chromosomal instability (CIN) by
missegregation of whole chromosomes [Thompson and Compton, 2011]. CIN and aneuploidy are
correlated with metastatic potential and contribute to resistance against chemotherapeutics in solid
tumors [Kuukasjarvi et al., 1997; McClelland et al., 2009; Swanton et al., 2009].
Daoy, LN-18 and SW 1783 cells were investigated with regard to their chromosome distribution to
evaluate their genetic stability and malignancy. Characteristic images obtained after cell preparation
and Giemsa staining for the determination of the chromosomal number in the corresponding cell line
are displayed in Figure 5.6.
A
B
C
10 µm
th
Figure 5.6: Metaphase chromosomes of Daoy (A), LN-18 (B) and SW 1783 cells (C) in the 7 , 18
respectively, stained with Giemsa.
th
st
and 81 passage,
The number of chromosomes was plotted against the frequency in which the respective number was
observed (Figure 5.7).
16
B
C
14
14
12
12
12
10
10
10
8
6
frequency (%)
14
frequency (%)
frequency (%)
16
16
A
8
6
8
6
4
4
4
2
2
2
0
30
103
40
50
60
70
80
90
100
0
30
number of chormosomes
40
50
60
70
80
90
100
0
30
40
50
60
70
80
90
100
number of chromosomes
number of chromosomes
th
th
st
Figure 5.7: Chromosome distribution of human Daoy (A), LN-18 (B) and SW 1783 cells (C) in the 7 , 18 and 81 passage,
respectively. For each passage the chromosomes of 50 well-spread metaphases were counted.
According to the information provided by the American Type Culture Collection (ATCC;
Manassas, VA) the Daoy and SW 1783 cell lines are hypertetraploid with a modal chromosome
number between 93 and 99. No karyotypical description was available for LN-18 cells. Whereas the
obtained modal number of 84 for the Daoy cell line was in good greement withliterature, the main
chromosome number of 56 determined for SW 1783 cells was different (Table 5.1). Unlike the Daoy
cells, SW 1783 and LN-18 cells revealed a comparatively uniform chromosomal distribution.
Table 5.1: Karyology of the used human brain tumor cell lines
cell line
Passage
obtained modal
chromosome number
range
modal chromosome
number (ATCC)
Daoy
7
84
65 - 96
93 - 99
LN-18
18
54
34 - 60
-
SW 1783
81
56
42 - 96
96
For all three cell lines only minor changes in the chromosomal distribution were recognized over
various in vitro passages (data not shown).
104
5.3.4
Chapter 5
Chemosensitivity against cytostatic drugs
The cell lines Daoy, LN-18 and SW 1783 were tested for their chemosensitivity against selected
anticancer drugs (Figure 5.8). The Daoy cell line was quite sensitive against all used cytostatics
(Figure 5.9). Mitoxantrone, vinblastine or paclitaxel (10 nM each) and methotrexate or etoposide at a
concentration of 1 µM yielded strong cytotoxic or cytocidal effects. Topotecan and doxorubicin
caused a cytostatic drug effect at 10 nM.
The LN-18 (Figure 5.10) and SW 1783 cells (Figure 5.11) tolerated topotecan and doxorubicin at
concentrations up to 10 nM. A concentration of 30 nM caused only a weak cytotoxic effect.
Etoposide was well tolerated up to 1 µM by LN-18 cells and produced a weak toxic effect on SW 1783
cells.
The effects of all used anticancer drugs on the proliferation of the three cell lines are summarized in
Figure 5.12. As topotecan, doxorubicin, mitoxantrone and etoposide were identified as ABCG2
substrates [Mo and Zhang, 2012], the results of the chemosensitivity studies suggest that both the
LN-18 and SW 1783 cell line express the ABCG2 protein in the wildtype.
105
Figure 5.8: Chemical structures of anticancer drugs used for chemosensitivity studies: Doxorubicin (Doxo), Mitoxantrone
(Mito), Etoposide (Etp), Topotecan (Topo), Paclitaxel (Pac), Vinblastine (Vbl), Methotrexate (MTX).
Chapter 5
140
3.0
80
40
1.5
20
0
1.0
-20
-40
absorbance A580
2.0
60
0.5
-60
0
20
40
60
80
2.5
80
60
2.0
1.5
20
0
0.0
100 120 140
1.0
-20
-40
0.5
-60
-80
0.0
0
20
40
60
80
100
3.0
3.0
120
100
C
40
1.5
20
1.0
(T-Co)/Co (%)
0
-20
0.5
-40
-60
0.0
20
40
60
80
100
120
2.5
60
2.0
40
20
1.5
0
(T-Co)/Co (%)
2.0
T/Ccorr. (%)
80
60
D
80
2.5
absorbance A580
T/Ccorr. (%)
100
0
120
1.0
-20
-40
0.5
-60
-80
0
20
40
60
80
0.0
100 120 140
absorbance A580
-80
B
100
40
(T-Co)/Co (%)
2.5
100
T/Ccorr. (%)
T/Ccorr. (%)
A
120
(T-Co)/Co (%)
3.0
120
absorbance A580
106
F
2.5
80
60
2.0
40
20
1.5
0
T/Ccorr. (%)
100
absorbance A580
-40
2.5
60
2.0
1.5
0
-20
0.5
-60
-80
80
20
1.0
-20
3.0
40
(T-Co)/Co (%)
T/Ccorr. (%)
100
(T-Co)/Co (%)
120
3.0
E
1.0
-40
0.5
-60
-80
0.0
0
20
40
60
80
100
0
120
absorbance A580
120
107
20
40
60
80
100
0.0
120
3.0
G
2.5
100
80
2.0
60
1.5
40
(T-Co)/Co (%)
20
absorbance A580
T/Ccorr. (%)
120
1.0
0
0.5
-20
-40
0
20
40
60
80
100
0.0
120
Figure 5.9: Chemosensitivity of Daoy cells against topotecan (A), mitoxantrone (B), doxorubicin (C), vinblastine (D) and
paclitaxel (E) at the following concentrations: 1 nM (filled circles), 10 nM (filled squares), 30 nM (filled triangles), 100 nM
(filled diamonds); F) Etoposide: 100 nM (filled circles), 1 µM (filled squares), 3 µM (filled triangles), 10 µM (filled diamonds);
G) Methotrexate: 1 µM (filled circles), 10 µM (filled squares), 30 µM (filled triangles), 100 µM (filled diamonds); vehicle
(open circles).
108
Chapter 5
120
120
3.0
2.0
1.5
40
1.0
0.5
0
20
40
60
80
1.5
40
1.0
0
-20
0.5
-40
-60
0.0
100 120 140
0
20
2.5
60
1.0
40
0.5
20
T/Ccorr. (%)
1.5
0.0
120
2.5
80
2.0
60
40
1.5
20
(T-Co)/Co (%)
2.0
absorbance A580
T/Ccorr. (%)
100
D
100
80
80
120
C
100
60
timie of incubation [h]
120
40
0
1.0
-20
-40
0.5
-60
0
0
20
40
60
80
0.0
100 120 140
-80
0
20
40
60
80
0.0
100 120 140
absorbance A580
20
60
20
(T-Co)/Co (%)
60
absorbance A580
T/Ccorr. (%)
80
2.0
80
absorbance A580
2.5
T/Ccorr. (%)
100
100
0
2.5
B
A
3.0
120
T/Ccorr. (%)
80
60
1.5
40
20
absorbance A580
2.0
-20
-40
20
40
60
80
0.0
100 120 140
2.5
20
2.0
0
(T-Co)/Co (%)
(T-Co)/Co (%)
0.5
60
40
1.0
0
3.0
80
1.5
-20
-40
1.0
-60
-80
0
absorbance A580
2.5
T/Ccorr. (%)
100
100
0
3.5
F
E
120
109
20
40
60
80
100
0.5
120
120
3.0
G
2.5
80
2.0
60
40
1.5
absorbance A580
T/Ccorr. (%)
100
(T-Co)/Co (%)
20
1.0
0
-20
0
20
40
60
80
100
0.5
120
Figure 5.10: Chemosensitivity of LN-18 cells against topotecan (A), mitoxantrone (B), doxorubicin (C), vinblastine (D) and
(open circles).
110
Chapter 5
120
A
2.0
100
2.0
1.2
40
0.8
20
-20
20
40
60
40
1.2
20
(T-Co)/Co (%)
0.4
0
1.6
60
0.8
0
-40
80
0.0
80 100 120 140 160
0
0.4
-20
-40
0
20
120
absorbance A580
1.6
80
T/Ccorr. (%)
100
60
(T-Co)/Co (%)
2.4
B
absorbance A580
T/Ccorr. (%)
120
40
60
0.0
80 100 120 140 160
1.0
120
1.2
D
C
100
100
1.0
0.8
0.4
0.8
60
40
0.6
20
0.4
0.2
0
0
20
40
60
0.0
80 100 120 140 160
(T-Co)/Co (%)
20
0
0.2
-20
-40
0
20
40
60
0.0
80 100 120 140 160
absorbance A580
40
absorbance A580
T/Ccorr. (%)
0.6
60
T/Ccorr. (%)
80
80
120
111
120
E
1.0
100
2.0
F
100
40
(T-Co)/Co (%)
20
0.4
0
0.2
-20
-40
0
20
40
60
0.0
80 100 120 140 160
80
60
1.2
40
0.8
20
0
0.4
-20
-40
0
absorbance A580
0.6
(T-Co)/Co (%)
60
T/Ccorr. (%)
0.8
absrobance A580
T/Ccorr. (%)
1.6
80
20
40
60
80
100
0.0
120
140
2.0
G
120
1.6
1.2
80
60
0.8
absorbance A580
T/Ccorr. (%)
100
40
0.4
20
0
0
20
40
60
80
100
0.0
120
Figure 5.11: Chemosensitivity of SW 1783 cells against topotecan (A), mitoxantrone (B), doxorubicin (C), vinblastine (D) and
(open circles).
112
Chapter 5
MTX
Etp
Pac
Vbl
Doxo
Mito
Topo
Daoy
1 nM
10 nM
30 nM
100 nM
1 nM
10 nM
30 nM
100 nM
1 nM
10 nM
30 nM
100 nM
1 nM
10 nM
30 nM
100 nM
1 nM
10 nM
30 nM
100 nM
100 nM
1 µM
3 µM
10 µM
1 µM
10 µM
30 µM
100 µM
LN-18 SW 1783
no effect
weak cytotoxic
cytostatic / cytotoxic
cytocidal
Figure 5.12: Summary of the different effects of the used anticancer drugs on the proliferation of Daoy, LN-18 and SW 1783
cells.
5.3.5
Characterization and growth kinetics of human brain tumor
cell lines in a subcutaneous tumor model in nude mice
5.3.5.1 In vivo growth kinetics
With respect to their potential use in a new orthotopic brain tumor model, Daoy, LN-18 and SW 1783
cells were investigated in vivo. The cells were subcutaneously injected into nude NMRI (nu/nu) mice.
Unfortunately, only the Daoy medulloblastoma cells were tumorigenic. Figure 5.13 shows the in vivo
growth kinetics of Daoy xenografts in consecutive passages when injected as cell suspension or
implanted as solid tumor pieces.
In general, Daoy tumors were hardly supplied with blood resulting in a long phase of tumor
formation (about 70 days) and a slow growth rate, especially in passages 0-2. Primarily in passage 3
the growth behavior changed. Take rates increased from 20% in passage 0 to 40-60% in the following
passages. Nevertheless, the growth rates of Daoy tumors have not been reproducible till the end of
this work. The body weight of animals was controlled as an indication of the general state of health.
For future work tumor pieces were cryopreserved and afterwards stored in liquid nitrogen.
113
mean tumor area [mm2]
250
200
150
Figure 5.13: Growth curves of subcutaneous Daoy
tumors over various passages after injection as cell
suspension (passage 0; passage 25 in cell culture;
3
black circles) or 1 mm tumor pieces (passages
1-3); passage 1: red triangles, passage 2: blue
squares, passage 3: green diamonds.
100
50
0
0
50
100
150
200
days after implantation
5.3.5.2 Histology
Figure 5.14 shows the morphology of a human Daoy medulloblastoma tumor grown in nude mice
after Masson-Goldner (MG) and haematoxylin-eosin (HE) staining. MG staining works with three
different stains allowing a differentiated visualization of tissues: Cell nuclei are displayed in dark
brown, cytoplasm and muscle fibers in red, connective tissue in green.
Attributes like anisomorphism and multiple prominent nucleoli highlight the malignant character of
the Daoy cell line. Mitotic activity, however, is low, resulting in a very slow in vivo growth.
A
B
Figure 5.14: MG (A) and HE (B) staining of a subcutaneous Daoy tumor (passage 0), 150 days after implantation
(BH-2 microscope, Olympus; objective 20x).
5.3.6
Investigations on ABCG2 induced cell lines
5.3.6.1 Western Blot analysis of wildtype and induced cell lines
Western Blot (WB) analysis and chemosensitivity tests (cf. section 5.3.4) revealed that LN-18 and
SW 1783 cells express the ABCG2 protein in the wildtype (Figure 5.15, lanes b and c). Besides these
two cell lines, additionally, Daoy and P388D1 cells were treated with increasing topotecan
114
Chapter 5
concentrations to yield sufficient ABCG2 transporter expression. Induction of BCRP overexpression
was monitored via WB analysis at different times after starting the topotecan treatment.
116
66
45
35
a
b
c
d
e
f
g
Figure 5.15: Immunological detection of ABCG2 expressed in various tumor cell lines. 20 µg of protein per cell lysate were
loaded on each lane. a: Daoy; b: LN-18; c: SW 1783; d: peqGold Protein Marker I; e: P388D1; f: P388D1/Topo (350 nM); g:
MCF-7/Topo (550 nM); numbers on the right designate masses of marker proteins in kDa.
After 10 weeks of treatment with topotecan up to a final concentration of 800 nM and 400 nM,
respectively, LN-18 and SW 1783 cells (in the following designated as LN-18/Topo and SW 1783/Topo
cells) show ABCG2 transporter expression (Figure 5.16) comparable to the standard cell line
MCF-7/Topo. The predicted molecular weight of the ABCG2 transporter monomer is 72 kDa. Band
analysis was performed with Quantity One (Bio-Rad) software and revealed a band of ~70 kDa for
MCF-7Topo cells (positive control) as well as for induced SW 1783 and LN-18 cells (Figure 5.16, lanes
d and g). Some diffuse bands were found at ~35 kDa, probably resulting from unspecific antibody
binding or proteolysis of the target BCRP protein.
To have access to a murine ABCG2 transporter system with regard to investigations in animal models,
the induction of ABCG2 was explored in the mouse cell line P388D1. However, although topotecan
concentrations as high as 1 µM were tolerated, there was no induction of ABCG2 transporter
expression detectable (Figure 5.16, lane b). By contrast, the Daoy cell line did not tolerate ascending
topotecan concentrations beyond a concentration of 30 nM.
115
140
80
50
30
a
b
c
d
e
f
g
h
i
j
Figure 5.16: Immunological detection of ABCG2 in tumor cell lines 10 weeks after starting BCRP induction by incubation
with topotecan. 20 µg of protein per cell lysate were loaded on each lane. a: Daoy/Topo (30 nM); b: P388D1/Topo (1 µM);
c: SW 1783; d: SW 1783/Topo (400 nM); e: peqGold Protein Marker I; f: LN-18; g: LN-18/Topo (800 nM); h: MCF-7/Topo
(550 nM); i: Kb-V1 (ABCG2-negative control); j: Biotinylated Protein Ladder (9-200 kDa, Cell Signaling); numbers on the right
designate masses of marker proteins in kDa.
For a rough estimation of the expression level in the newly induced cell lines in comparison to the
standard MCF-7/Topo cell line, the specular optical densities (OD) of the bands around 70 kDa were
calculated after calibration by the Quantity One software for the GS-710 Imaging Densitometer
(Table 5.2). The three topotecan treated cells types seemed to express the ABCG2 protein to a
similar extent.
Table 5.2: Comparison of Western Blot band analysis for induced cell lines.
lane
cell line
calculated MW [kDa] band OD
d
SW 1783/Topo (400 nM)
69.24
2.07
g
LN-18/Topo (800 nM)
68.36
2.15
h
MCF-7/Topo (550 nM)
70.02
2.00
5.3.6.2 Determination of ABCG2 overexpression by flow cytometry
In addition to the Western Blot analysis, an indirect flow cytometric FACS assay was performed to
compare the ABCG2 transporter expression in LN-18 and SW 1783 wildtype cells to the expression in
the corresponding topotecan treated cells. By analogy with the Western Blot analysis, an increased
ABCG2 transporter expression was determined for LN-18/Topo (Figure 5.17) and SW 1783/Topo cells
(Figure 5.18) compared to the wt cells. The quantification was based on the differences between the
corresponding geometrical mean (GeoMean) values. Autofluorescence and the fluorescence of the
cells under treatment with the secondary antibody alone were used as negative controls.
116
Chapter 5
A
B
500
GeoMean
400
300
200
100
0
lu
of
t
au
c
es
en
ce
or
se
co
n
AB
ry
da
wt
18
LN
LN
p
/To
18
o
(
0
80
nM
)
-
Figure 5.17: A) ABCG2 transporter expression determined by flow cytometric analysis on human LN-18 wildtype (blue) and
LN-18/Topo (800 nM) cells (red); autofluorescense (black), secondary antibody (green). B) GeoMean values (+ SEM) of
LN-18 wild type and topotecan induced cells calculated from 2 independent experiments.
Only for SW 1783 cells, marginal non-specific binding of the secondary antibody (goat anti-rat IgG-PE)
was detectable. The GeoMean value of LN-18 cells was doubled by the treatment with topotecan,
whereas the GeoMean of SW 1783/Topo cells was three times as high as for the wildtype cells. The
induced SW 1783 cells contained a subpopulation, which was not further influenced by the
treatment with topotecan (Figure 5.18 (A), red graph).
1400
A
B
1200
GeoMean
1000
800
600
400
200
0
lu
of
t
au
c
es
or
en
ce
ry
se
co
a
nd
AB
SW
17
83
wt
83
SW
0n
op
/T
o
M)
0
(4
17
Figure 5.18: A) Graphical illustration of ABCG2 transporter expression via flow cytometric analysis in human SW 1783
wildtype (blue) and in SW 1783/Topo (400 nM) cells (red); autofluorescense (black), secondary antibody (green).
B) GeoMean values (+ SEM) of SW 1783 wild type and topotecan induced cells calculated from 2 independent experiments.
5.3.6.3 Hoechst 33342 assay using ABCG2 induced cancer cells
The Hoechst 33342 assay was performed with the ABCG2 overexpressing brain tumor cells to
determine the transporter inhibition by a selection of synthesized modulators according to the
117
standard procedure described for MCF-7/Topo cells in section 3.3.5.1 using fumitremorgin C as the
reference substance.
Table 5.3: Inhibition of ABCG2 by selected modulators in the Hoechst 33342 assay using three
different BCRP overexpressing cell lines.
MCF-7/Topoa
Compound
SW 1783/Topob
LN-18/Topob
IC50 [nM]
Imax (%)
IC50 [nM]
Imax (%)
IC50 [nM]
Imax (%)
FTC
731 ± 92
100
842 ± 101
114 ± 19
1,129 ± 203
113 ± 12
12 (UR-COP228)
591 ± 87
109 ± 8
625 ± 52
117 ± 12
633 ± 255
105 ± 2
32 (UR-COP269)
59 ± 11
101 ± 5
68 ± 21
108 ± 6
71 ± 18
104 ± 13
33 (UR-COP272)
46 ± 1
111 ± 9
40 ± 9
112 ± 22
60 ± 16
109 ± 10
36 (UR-MB83-2)
168 ± 81
96 ± 1
173 ± 98
105 ± 18
257 ± 174
86 ± 9
41 (UR-MB108-4)
64 ± 3
95 ± 3
48 ± 10
101 ± 2
120 ± 42
103 ± 9
a
Mean values (+ SEM) were calculated from two to three independent experiments performed in sextuplicate;
values (+ SEM) were calculated from two independent experiments performed in triplicate.
b
Mean
As shown in Table 5.3 and displayed in Figure 5.19 for compounds 32 and 33, the obtained IC50
values and the maximal inhibitory effects are almost identical in the Hoechst 33342 assay,
independent from the used cells. Therefore, these three cell lines are considered to be useful for the
study of ABCG2 modulators in the H33342 assay.
120
A
100
100
120
80
60
40
20
80
60
40
20
0
0
0.001
B
0.01
0.1
1
concentration [µM]
10
100
0.001
0.01
0.1
1
concentration [µM]
10
100
Figure 5.19: Inhibition of ABCG2 by compounds 32 (A) and 33 (B) using MCF-7/Topo (filled circles), SW 1783/Topo (filled
triangles) and LN-18/Topo cells (filled squares).
118
Chapter 5
5.3.6.4 Chemosensitivity of LN-18/Topo cells against selected cytostatics
The effect of mitoxantrone, doxorubicin (ABCG2 substrates) and vinblastine on the proliferation of
ABCG2 overexpressing LN-18/Topo cells were investigated by means of a kinetic chemosensitivity
assay. The results of the drug effects on wildtype (cf. section 5.3.4) and ABCG2 induced LN-18 cells
are shown in Figure 5.20 and Figure 5.21.
The treatment of LN-18 wildtype cells with mitoxantrone at concentrations of 30 nM and 100 nM,
respectively, caused a strong cytocidal drug effect. LN-18/Topo cells, however, were resistant against
this ABCG2 substrate at concentrations up to 100 nM.
In case of doxorubicin, a moderate substrate of BCRP, the differences between the effects on the two
cell types were less significant as for mitoxantrone. Nevertheless, a high concentration of 100 nM
resulted in a strong cytotoxic effect on LN-18 wt cells and only in a weak toxic effect on the ABCG2
induced cells.
As expected, the chemosensitivity against vinblastine (P-gp substrate) did not differ between the
wildtype and LN-18/Topo cells (data not shown).
2.5
A
1.4
1.2
100
60
20
1.0
0
1.0
T/Ccorr. (%)
1.5
absorbance A580
T/Ccorr. (%)
2.0
80
40
(T-Co)/Co (%)
B
120
100
80
0.8
60
0.6
absorbance A580
120
40
-20
0.4
0.5
-40
20
0.2
-60
0
20
40
60
80
100
0.0
120
0
0
20
40
60
0.0
80 100 120 140 160
Figure 5.20: Chemosensitivity of LN-18 wt (A) and LN-18/Topo (B) cells against mitoxantrone at the following
concentrations: 10 nM (filled circles), 30 nM (filled squares), 100 nM (filled triangles); vehicle (open circles).
2.5
1.4
A
120
119
B
120
1.2
2.0
100
100
1.0
60
1.0
0.8
60
0.6
40
40
20
0.4
0
0.2
absorbance A580
1.5
T/Ccorr. (%)
T/Ccorr. (%)
80
absorbance
80
0.5
20
0
0
20
40
60
80
0.0
100 120 140
time of incubation
-20
0
20
40
60
0.0
80 100 120 140 160
Figure 5.21: Chemosensitivity of LN-18 wt (A) and LN-18/Topo (B) cells against doxorubicin at the following concentrations:
10 nM (filled circles), 30 nM (filled squares), 100 nM (filled triangles); 300 nM (filled diamonds); vehicle (open circles).
5.3.7
ABC transporter expression in HMEC-1 cells
Immortalized human microvascular endothelial cells (HMEC-1) were investigated for ABCB1, ABCG2
and ABCC1 expression by means of a FACS assay as described in section 5.2.10 (GeoMean values cf.
Table 5.4). The endothelial cells (EC) revealed moderate expression of MRP1 (ABCC1) and ABCG2,
whereas P-gp (ABCB1) was not detected.
The test for ABC transporter expression via flow cytometric analysis was performed only once due to
a very slow in vitro growth of the cells. The results have to be confirmed by additional experiments,
nevertheless they are in a good agreement with recently published data [Eilers et al., 2008].
Eilers et al. monitored the expression of ABC transporters in primary EC from different human
tissues, including dermal microvascular endothelial cells. The authors report on a significant ABCC1
transporter expression in all investigated ECs, whereas P-gp levels were lowest for microvessel
derived ECs.
ABCB1 ABCC1 ABCG2
autofluorescence
72.4
72.1
72.6
sec. AB
122.0
79.0
89.9
prim. + sec. AB
118.4
147.0
129.3
Table 5.4: GeoMean values for ABC transporter expression in
HMEC-1 cells determined by flow cytometry. Background
fluorescence of HMEC-1 cells and non-specific binding of the
secondary antibody were detected and used as negative
controls.
120
Chapter 5
Figure 5.22: Flow cytometric analysis of ABCG2 (A),
ABCC1 (B) and ABCB1 (C) transporter expression in
HMEC-1 cells; primary and secondary antibody (red),
secondary antibody (green), autofluorescense
(black).
5.4
121
Summary
The malignant brain tumor cells Daoy, LN-18 and SW 1783 were investigated with regard to their
morphology, in vitro growth kinetics, chemosensitivity against anticancer drugs, chromosome
distribution and ABC transporter expression. All cell lines show pronounced features of high
malignancy and genetic instability. Chemotoxicity studies revealed a high sensitivity of the Daoy cell
line against almost all used cytotstatics whereas LN-18 and SW 1783 cells were at least partially
resistant, for instance, against topotecan, doxorubicin and etoposide. Western Blot analysis and
chemosensitivity tests indicated a slight ABCG2 transporter expression in LN-18 and SW 1783
wildtype cells. Treatment of these cells with increasing concentrations of topotecan yielded two
BCRP overexpressing cell lines which are comparable to the standard MCF-7/Topo cell line and are
therefore suitable for the use in in vitro assays.
The Daoy medulloblastoma cells were successfully established as subcutaneously growing tumors in
nude mice, whereas LN-18 and SW 1783 cells were not tumorigenic under the same conditions.
Results of FACS investigations on HMEC-1 cells suggest expression of the ABCC1 transporter in these
endothelial cells.
122
5.5
Chapter 5
References
Abbott, N. J., Patabendige, A. A., Dolman, D. E., et al. Structure and function of the blood-brain
barrier. Neurobiol. Dis. 2010, 37(1), 13-25.
Ades, E. W., Candal, F. J., Swerlick, R. A., et al. HMEC-1: establishment of an immortalized human
microvascular endothelial cell line. J. Invest. Dermatol. 1992, 99(6), 683-690.
Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein
utilizing the principle of protein-dye binding. Anal. Biochem. 1976, 72, 248-254.
Cohen, M. H., Shen, Y. L., Keegan, P., et al. FDA drug approval summary: bevacizumab (Avastin) as
treatment of recurrent glioblastoma multiforme. Oncologist 2009, 14(11), 1131-1138.
Cook, L. J. and Freedman, J. Brain Tumors. The Rosen Publishing Group, Inc., New York, USA, 2012.
Dolecek, T. A., Propp, J. M., Stroup, N. E., et al. CBTRUS statistical report: primary brain and central
nervous system tumors diagnosed in the United States in 2005-2009. Neuro Oncol. 2012, 14
Suppl 5, v1-49.
Eilers, M., Roy, U. and Mondal, D. MRP (ABCC) transporters-mediated efflux of anti-HIV drugs,
saquinavir and zidovudine, from human endothelial cells. Exp. Biol. Med. (Maywood) 2008,
233(9), 1149-1160.
Ferlay, J., Shin, H. R., Bray, F., et al. Estimates of worldwide burden of cancer in 2008: GLOBOCAN
2008. Int. J. Cancer. 2010, 127(12), 2893-2917.
Ferlay, J., Steliarova-Foucher, E., Lortet-Tieulent, J., et al. Cancer incidence and mortality patterns in
Europe: estimates for 40 countries in 2012. Eur. J. Cancer 2013, 49(6), 1374-1403.
Huse, J. T. and Holland, E. C. Targeting brain cancer: advances in the molecular pathology of
malignant glioma and medulloblastoma. Nat. Rev. Cancer 2010, 10(5), 319-331.
Jarzyna, P. Preclinical investigations on the effect of the human hyaluronidase Hyal-1 on growth and
metastasis of human colon carcinoma. PhD thesis, University of Regensburg, Germany, 2007.
Kuukasjarvi, T., Karhu, R., Tanner, M., et al. Genetic heterogeneity and clonal evolution underlying
development of asynchronous metastasis in human breast cancer. Cancer Res. 1997, 57(8),
1597-1604.
Laemmli, U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4.
Nature 1970, 227(5259), 680-685.
Lee, G., Dallas, S., Hong, M., et al. Drug transporters in the central nervous system: brain barriers and
brain parenchyma considerations. Pharmacol. Rev. 2001, 53(4), 569-596.
Louis, D. N., Ohgaki, H., Wiestler, O. D., et al. The 2007 WHO classification of tumours of the central
nervous system. Acta Neuropathol. 2007, 114(2), 97-109.
Martin-Banderas, L., Holgado, M. A., Venero, J. L., et al. Nanostructures for drug delivery to the brain.
Curr. Med. Chem. 2011, 18(34), 5303-5321.
123
McClelland, S. E., Burrell, R. A. and Swanton, C. Chromosomal instability: a composite phenotype that
influences sensitivity to chemotherapy. Cell Cycle 2009, 8(20), 3262-3266.
Mo, W. and Zhang, J. T. Human ABCG2: structure, function, and its role in multidrug resistance. Int. J.
Biochem. Mol. Biol. 2012, 3(1), 1-27.
Mulisch, M. and Welch, U. Romeis Mikroskopische Technik. Spektrum Akademischer Verlag,
Heidelberg, Germany, 2010.
Müller, C. New approaches to the therapy of glioblastoma: investigations on RNA interference,
kinesin Eg5 and ABCB1/ABCG2 inhibition. PhD thesis, University of Regensburg, Germany,
2007.
Newlands, E. S., Stevens, M. F., Wedge, S. R., et al. Temozolomide: a review of its discovery, chemical
properties, pre-clinical development and clinical trials. Cancer Treat. Rev. 1997, 23(1), 35-61.
Orthmann, A., Zeisig, R., Suss, R., et al. Treatment of experimental brain metastasis with MTOliposomes: impact of fluidity and LRP-targeting on the therapeutic result. Pharm. Res. 2012,
29(7), 1949-1959.
Panigrahi, M., Das, P. K. and Parikh, P. M. Brain tumor and Gliadel wafer treatment. Indian J. Cancer
2011, 48(1), 11-17.
Pardridge, W. M. The blood-brain barrier and neurotherapeutics. NeuroRx. 2005, 2(1), 1-2.
Pertschuk, D., Cloughesy, T. F., Chang, S. M., et al. Ascending dose trials of the safety and tolerability
of Toca 511, a retroviral replicating vector encoding cytosine deaminase, in patients with
recurrent high-grade glioma. J. Clin. Oncol. 2012, 30(15).
Robbins, J. M., Dickinson, P. J., York, D., et al. Evaluation of Delivery of Retroviral Replicating Vector,
Toca 511, in Spontaneous Canine Brain Tumor. Neuro Oncol. 2012, 14, 48-48.
Rooney, D. E. and Czepulkowski, B. H. Human Cytogenetics - A Practical Approach. Oxford University
Press, USA, 1992.
Rousselle, C., Smirnova, M., Clair, P., et al. Enhanced delivery of doxorubicin into the brain via a
peptide-vector-mediated strategy: saturation kinetics and specificity. J. Pharmacol. Exp. Ther.
2001, 296(1), 124-131.
Saito, R. and Tominaga, T. Convection-enhanced delivery: from mechanisms to clinical drug delivery
for diseases of the central nervous system. Neurol. Med. Chir. (Tokyo) 2012, 52(8), 531-538.
Sekine, T., Cha, S. H. and Endou, H. The multispecific organic anion transporter (OAT) family. Pflugers
Arch. 2000, 440(3), 337-350.
Shih, T. and Lindley, C. Bevacizumab: an angiogenesis inhibitor for the treatment of solid
malignancies. Clin. Ther. 2006, 28(11), 1779-1802.
Stamatovic, S. M., Keep, R. F. and Andjelkovic, A. V. Brain endothelial cell-cell junctions: how to
"open" the blood brain barrier. Curr. Neuropharmacol. 2008, 6(3), 179-192.
Stupp, R., Mason, W. P., van den Bent, M. J., et al. Radiotherapy plus concomitant and adjuvant
temozolomide for glioblastoma. N. Engl. J. Med. 2005, 352(10), 987-996.
124
Chapter 5
Swanton, C., Nicke, B., Schuett, M., et al. Chromosomal instability determines taxane response. Proc.
Natl. Acad. Sci. U S A 2009, 106(21), 8671-8676.
Takahashi, M. Farbatlas der onkologischen Zytologie. Perimed-Fachbuch-Verlag-Ges., Erlangen,
Germany, 1987.
Thompson, S. L. and Compton, D. A. Chromosomes and cancer cells. Chromosome Res. 2011, 19, 433444.
Tosi, G., Bortot, B., Ruozi, B., et al. Potential Use of Polymeric Nanoparticles for Drug Delivery Across
the Blood-Brain Barrier. Curr. Med. Chem. 2013, 20(17), 2212-2225.
Walker, M. D. and Hurwitz, B. S. BCNU (1,3-bis(2-chloroethyl)-1-nitrosourea; NSC-409962) in the
treatment of malignant brain tumor--a preliminary report. Cancer Chemother. Rep. 1970,
54(4), 263-271.
Wen, P. Y. and Kesari, S. Malignant gliomas in adults. N. Engl. J. Med. 2008, 359(5), 492-507.
Chapter 6
6
Towards an ATPase assay for the
human ABCG2 transporter
6.1
Introduction
ABC transporter-mediated drug transport across membranes against a concentration gradient is
driven by energy provided by the hydrolysis of ATP. Generated inorganic phosphate (Pi) is
proportional to the ATPase activity of the transporter and can, in principle, be quantified by a
colorimetric assay on cell membranes. Membrane preparations of recombinant baculovirus infected
Sf9 cells are widely used to investigate the interaction of compounds with different members of the
ABC transporter family [Germann et al., 1990; Sarkadi et al., 1992]. Activation as well as inhibition of
ABC transporters can be investigated by measuring ATPase activity: Compounds that stimulate
ATPase activity are generally considered substrates of the transporter whereas inhibitors decrease
stimulated activity.
An ABCG2 ATPase assay harbors the potential of providing information on the mechanism of action
of the synthesized ABCG2 modulators described in Chapter 3.
In the present work, an ATPase assay for the human ABCG2 transporter was established using
membrane preparations of Sf9 insect cells. The gene encoding ABCG2 was cloned into the pVL1392
vector, expressed in Sf9 cells and an experimental protocol for the ATPase assay was set up.
6.2
Materials and Methods
6.2.1
Materials
The pcDNA5/FRT-ABCG2 and pcDNA5/FRT-ABCB1 plasmids were a kind gift of Prof. Dr. Deanna
Kroetz (Departments of Biopharmaceutical Sciences and Pharmaceutical Chemistry, University of
California, San Francisco, CA). Phusion® High-Fidelity DNA Polymerase (2 U/µL) and the
126
Chapter 6
corresponding HF Buffer (5x), deoxynucleotide solution mix (dNTPs), the restriction enzymes XbaI
and NotI-HF™, Antarctic Phosphatase, purified BSA (100x) and NEBuffer 4 for digestion, 6x Loading
Dye, 100 bp DNA Ladder, 2-log DNA Ladder (0.1-10.0 kb) as well as T4 DNA Ligase (6 Weiss U/μL) and
T4 DNA Ligase Reaction Buffer (10x) were all purchased from New England Biolabs (NEB; Ipswich,
MA). All used primers were from Eurofins MWG Synthesis GmbH (Ebersberg, Germany). Melting
temperatures (Tm) of the primers for estimating annealing temperatures were calculated
using the online Tm Calculator of Thermo Scientific (www.thermoscientificbio.com/webtools/tmc/).
Agarose and gel chambers of the perfectBlue™ Mini S gel system were obtained from
Peqlab (Erlangen, Germany). QIAquick Gel Extraction, QIAquick PCR Purification, QIAprep Spin
Miniprep and QIAGEN Plasmid Maxi Kits were from QIAGEN (Hilden, Germany).
TAE buffer was made of 48.4 g/L Tris base (usb, Cleveland, OH), 11.42 mL glacial acetic acid and
20 mL EDTA solution (0.5 M, pH 8.0) in 1000 mL purified water. Ampicillin was obtained from Sigma
(Munich, Germany).
PCR reactions were performed in a Mastercycle gradient Thermocycler (Eppendorf, Hamburg,
Germany) and sequencing was performed by Entelechon (Bad Abbach, Germany) or Eurofins MWG
Synthesis GmbH.
RIPA buffer for cell lysis was made of 20 mM Tris, 200 mM NaCl, 1 mM EDTA, 1% (v/v) Triton X-100,
5 mM K2HPO4. The pH was adjusted to 7.8 and 100 mL of buffer were supplemented with a protease
inhibitor tablet (SIGMAFAST™ Tablets; Sigma).
All used antibodies were from Santa Cruz Biotechnology (Heidelberg, Germany).
3-Morpholinopropane-1-sulfonic acid (MOPS), sodium orthovanadate and EGTA for ATPase reaction
buffer were purchased from Sigma. If not otherwise stated, chemicals (p.a. quality) were from Merck
(Darmstadt, Germany). Purified water (Milli-Q system, Millipore, Eschborn, Germany) was used.
6.2.2
Transformation of E. coli
LB and recovery (SOC) media, selective agar plates and competent bacteria E. coli (TOP10 strain)
were prepared as described before [Pop, 2010].
For chemical transformation, 200 μL of a suspension of competent E. coli cells were thawed on ice
and approximately 1 ng of the plasmid DNA of interest was added prior to incubation on ice (30 min).
Cells were heat-shocked (42 °C) for 90 sec and subsequently put on ice for 2 min. 1 mL of prewarmed SOC medium was added and the bacteria were incubated for 45 min at 37 °C under shaking
(200 rpm). The suspension was centrifuged for 5 min at 2,300 g. The bacterial pellet was
resuspended in 20-50 μL of the residual medium and plated on ampicillin-containing (100 µg/mL)
Towards an ATPase assay for the human ABCG2 transporter
127
agar plates for selection, as the used vectors included an ampicillin resistance gene. Plates were
incubated overnight at 37 °C. Resistant colonies were picked the next day and used for overnight
cultures in LB medium supplemented with 100 µg/mL ampicillin (amp-LB medium).
6.2.3
General procedures for preparation of plasmid DNA
6.2.3.1 Mini- and Maxi-Prep
For quick purification of up to 20 µg of high-copy plasmid DNA from 5 mL overnight culture of E. coli
(in amp-LB medium), QIAprep Spin Miniprep Kit was used according to the manual for
microcentrifugation.
For large scale preparations of DNA, 250 mL of amp-LB medium were inoculated with 250 µL of
bacterial overnight culture suspension. QIAGEN Plasmid Maxi Kit was used for pure DNA isolation
according to the manufacturer’s instructions.
Usually, a 1:20 dilution of Mini-Prep or Maxi-Prep DNA was prepared and DNA concentration was
determined at a BioPhotometer (Eppendorf; Hamburg, Germany).
6.2.3.2 Restriction enzyme digestion and dephosphorylation of plasmid ends
For restriction enzyme analysis of plasmid DNA and the subcloning of PCR products, the enzymes
NotI-HF™ (20 U/µL) and XbaI (20 U/µL) were used. Double digestion was performed in a final volume
of 50 µL of millipore water containing 5 µL NEBuffer 4 (10x), 0.5 µL BSA (100x), 1 μL of each enzyme
and 2 μg of DNA or pVL1392 Baculovirus Transfer Vector (5 µg in 50 µL; BD Biosciences, Heidelberg,
Germany) for 2 h at 37 °C in an Eppendorf reaction vessel. Enzymes were heat-inactivated for
additional 20 min at 65 °C.
In order to avoid self-ligation, the DNA ends of the digested pVL1392 vector were dephosphorylated
by incubating for 1 h at 37 °C with 1/10 volume of Antarctic Phosphatase (10x) followed by 5 min
heat-inactivation at 65 °C.
6.2.3.3 Agarose gel electrophoresis
Agarose gels (1% or 3%) were prepared by dissolving 0.5 g or 1.5 g of agarose under heating and
stirring in 50 mL of TAE buffer in a microwave. To visualize DNA, 1 μL of an ethidium bromide
(Janssen Chimica; Beerse, Belgium) solution (10 mg/mL) was added. Gelling of the warm agarose
solution was performed in a gel chamber for 30 min using a comb as placeholder to create gel
pockets.
128
Chapter 6
For analytical gels usually 5 μL of DNA products mixed with 1 μL of 6x DNA loading dye were applied.
For quantitative gels after restriction enzyme digestion volumes up to 50 μL were pipetted into gels
with larger pockets.
The 2-log DNA ladder (0.1-10.0 kb; 7-10 μL of) was used as reference to estimate the length of PCR or
digestion products. Electrophoresis (in TAE buffer) was performed at 150 V (ca. 90 min) until the
track dye has moved 2/3 of the gel length. After removing the gel tray from the electrophoresis
chamber, DNA bands were visualized by illumination with UV light at 254 nm (Gel Doc 2000; Bio-Rad
Laboratories, Munich, Germany) using Quantity One (Bio-Rad) software for data analysis.
6.2.3.4 Purification of PCR products and recovery of DNA fragments from
agarose gels
DNA from PCRs was desalted and the polymerase was removed with the QIAquick PCR Purification
Kit. DNA fragments after restriction enzyme digestion were first separated by agarose gel
electrophoresis. Afterwards, bands with the correct sizes were excised from the gel under UV light
(λex: 254 nm) and extracted with the QIAquick Gel Extraction kit according to the manual.
6.2.4
Preparation of the S-ABCG2 construct via sequential overlap
extension PCR
Primers were designed to provide two different PCR products, the signal peptide sequence (ATG AAG
ACG ATC ATC GCC CTG AGC TAC ATC TTC TGC CTG GTA TTC GCC), a domain of 48 bp necessary for a
correct transporter expression in Sf9 cells, and the ABCG2 template, with a region of common
sequence. The two overlapping fragments were fused in a subsequent PCR amplification to obtain
the extended ABCG2 template (Figure 6.1).
In PCR 1a, the DNA sequence of the signal peptide (S) was amplified. A plasmid containing the signal
peptide sequence and the FLAG (F) epitope was used as template with the sense primer SF_NotIfor
encoding the NotI restriction site followed by the first 18 bp of the signal peptide and the antisense
primer Signal_rev comprising the final part (18 bp) of the signal peptide sequence.
In PCR 1b the cDNA of the ABCG2 receptor was amplified using the pcDNA5/FRT-ABCG2 vector as
template. PCR 1b was performed with the sense primer ABCG2_for encoding the final 18 bp of the
signal peptide followed by the initial 23 bp of the ABCG2 receptor and the antisense primer
ABCG2_XbaI_rev comprising 31 bp of the C-terminal sequence of the ABCG2 receptor and the XbaI
restriction site.
129
Due to their terminal complementarity, PCR products 1a and 1b, templates for PCR 2, will overlap
and subsequently be extended.
PCR 1a:
S
FLAG
pcDNA3.1(+)-SF-mH4R-His6
NotI
S
PCR 1a product
PCR 1b:
ABCG2
pcDNA/FRT-ABCG2
S
ABCG2
XbaI
PCR 1b product
PCR 2:
PCR 1a product
NotI
S
S
ABCG2
XbaI
PCR product 1b
NotI
S
ABCG2
XbaI
PCR 2 product
Figure 6.1: Strategy for the construction of the S-ABCG2 insert via overlap extension PCR. In separate PCR reactions (PCR 1a
and 1b), two fragments of the target sequence are amplified. PCR 1a product contains the complete signal peptide
sequence (S), whereas PCR 1b product comprises the complete sequence of the ABCG2 transporter and the final 18 bp of
the signal peptide. In PCR 2 the two intermediate products with terminal complementarity form a new template DNA,
annealing at the overlapping 18 bp of the signal peptide sequence. Primers are displayed as blue arrows.
130
Chapter 6
6.2.4.1 PCR 1a for the S-ABCG2 construct
The template pcDNA3.1(+)-SF-mH4R-His6 containing the signal peptide sequence was kindly provided
by Dr. Uwe Nordemann (Institute of Pharmacy, University of Regensburg).
sense primer SF_NotIfor:
5’- AGC TGC GGC CGC ATG AAG ACG ATC ATC GCC - 3’
NotI
antisense primer Signal_rev:
signal peptide
5’- GGC GAA TAC CAG GCA GAA - 3’
The PCR reaction mixture contained 1 ng template DNA, 0.5 µL (1 U) of Phusion® High−Fidelity DNA
polymerase, 10 μL of 5x HF buffer, 1 µL dNTP Mix (10 mM), primers (0.5 μM each) and water up to a
volume of 50 μL.
PCR reaction was performed with a gradient at 3 different
1000
annealing temperatures:
(1) initial denaturation
98 °C, 30 sec
(2) denaturation
98 °C, 10 sec
(3) annealing
55 °C, 60 °C or 65 °C, 15 sec
(4) extension
72 °C, 5 sec
(5) final extension
72 °C, 5 min
(6) hold
4 °C
a
b
c
500
200
100
60
Steps 2-4 were repeated 30 times.
Figure 6.2 shows the expected product (60 bp) of PCR 1a on a 3%
agarose gel. All three annealing temperatures (lanes a-c) are
suitable for this PCR reaction with a preference for 55 °C. The
PCR product (lane a) was purified using the QIAquick PCR
Purification Kit without further separation from possible side
products. The DNA concentration was determined.
Figure 6.2: Agarose gel (3%)
electrophoresis of PCR 1a product;
numbers indicate the sizes of the
bands in base pairs (bp); Lanes a-c
represent PCR products at 3 different
annealing temperatures: 55 °C (a),
60 °C (b), 65°C (c).
131
6.2.4.2 PCR 1b for the S-ABCG2 construct
PCR reaction mixture was the same as described for PCR 1a. 1 ng of the pcDNA5/FRT-ABCG2 vector
was used as template.
sense primer ABCG2_for:
5’- TTC TGC CTG GTA TTC GCC ATG TCT TCC AGT AAT GTC GAA GT - 3’
signal peptide
ABCG2
antisense primer ABCG2_XbaI_rev:
5’- ATA TCT AGA TTA AGA ATA TTT TTT AAG AAA TAA CAA TTT C - 3’
XbaI
ABCG2
PCR reaction was performed with a gradient at 3 different
10000
annealing temperatures:
98 °C, 30 s
(2) denaturation
98 °C, 10 s
(3) annealing
52 °C, 57 °C or 62 °C, 15 sec
(4) extension
72 °C, 40 s
(5) final extension
72 °C, 5 min
(6) hold
4 °C
Figure 6.3 shows the expected ABCG2 construct (1992 bp)
of PCR 1b on a 1% agarose gel. All three annealing
temperatures (lanes a-c) are appropriate for this PCR
3000
2000
1500
1992
1000
a
b
c
Figure
6.3:
Agarose
gel
(1%)
electrophoresis of PCR 1b product; lanes
a-c indicate the 3 different annealing
temperatures: 52 °C (a); 57 °C (b); 62°C (c).
reaction. The PCR product (lane a) was purified using the
QIAquick PCR Purification Kit without further separation.
The DNA concentration was determined.
6.2.4.3 PCR 2 for the S-ABCG2 construct
The annealed products of PCR 1a and PCR 1b served as template for the amplification of the new
insert S-ABCG2 flanked by the restriction sites NotI (5’ end) and XbaI (3’ end) in PCR 2.
The sense primer NotI_for (5’- AGC TGC GGC CGC - 3’) encoding the NotI restriction site was used.
ABCG2_XbaI_rev from PCR 1b served as antisense primer.
132
Chapter 6
Due to the extremely different base pair lengths of PCR products 1a (60 bp) and 1b (1992 bp) three
different PCR reactions were performed varying the ratio of concentrations of template 1a and 1b.
Figure 6.4 shows the results of PCR 2 fusing PCR 1a product with product 1b. The template
concentrations of PCR reactions shown in lanes a and b were calculated according the DNA length
ratio of the PCR products. All other reagents were used as described above in a total volume of 50 µL
under cycling conditions according to PCR 1b with an annealing temperature of 57 °C. QIAquick PCR
Purification Kit was used for purification.
10000
3000
2000
2034
1000
a
6.2.5
b
c
Figure 6.4: Agarose gel (1%) electrophoresis of PCR product 2
(2034 bp); lanes a-c indicate 3 different ratios: a: 0.03 ng (PCR 1a
product) + 1 ng (PCR 1b product), b: 0.3 ng (1a) + 10 ng (1b), c: 1 ng
(1a) + 10 ng (1b).
Preparation of the S-ABCB1 construct
Overlap extension PCR performed in the same way as for the S-ABCG2 construct was not successful
for the ABCB1 transporter. In a second approach, the signal peptide sequence was completely
inserted in the sense primer, flanked by the NotI restriciton site and the first 19 bp of the ABCB1
transporter (ordered as unmodified DNA oligo from Eurofins MWG Synthesis GmbH).
ABCB1_XbaI_rev served as antisense primer comprising the XbaI restriction site and the final 17 bp of
the ABCB1 transporter.
sense primer ABCB1_for(+S):
5‘- AGC TGC GGC CGC ATG AAG ACG ATC ATC GCC CTG AGC TAC ATC TTC TGC CTG GTA TTC GCC ATG
GAT CTT GAA GGG GAC C - 3‘
antisense primer ABCB1_XbaI_rev:
5’- ATA TCT AGA TCA CTG GCG CTT TGT TC - 3’
XbaI
ABCB1
133
Two different PCR strategies were chosen:
1. PCR 3a reaction was performed with a gradient at 3 different annealing temperatures:
98 °C, 30 s
(2) denaturation
98 °C, 10 s
(3) annealing
57 °C, 62 °C or 67 °C, 15 s
(4) extension
72 °C, 2 min
(5) final extension
72 °C, 5 min
(6) hold
4 °C
2. Due to the length and the associated high melting temperature of the sense primer
ABCB1_for(+S), additionally, PCR 3b reaction was performed in two cycles with an initial low
and a following high annealing temperature:
98 °C, 30 s
(2) denaturation
98 °C, 10 s
(3) annealing
60 °C, 15 s
(4) extension
72 °C, 2 min
(5) denaturation
98 °C, 10 s
(6) annealing
70 °C, 15 s
(7) extension
72 °C, 2 min
(8) Final extension
72 °C, 5 min
(9) hold
4 °C
a
3909
b
c
d
5000
4000
3000
1000
Figure 6.5: Agarose gel (1%) electrophoresis of PCR products 3a
and 3b; lanes a-c indicate PCR 3a at three different annealing
temperatures: 57 °C (a), 62 °C (b), 67°C (c); lane d indicates
PCR 3b.
134
Chapter 6
Figure 6.5 shows the expected product (3909 bp) of PCR 3a and PCR 3b on a 1% agarose gel. All three
annealing temperatures of PCR 3a (lanes a-c) are suitable for this PCR reaction with a preference for
57 °C. PCR 3b (lane d) with the two different annealing cycles did not give the desired product. PCR
product of lane a was purified and the DNA concentration was determined.
6.2.6
Subcloning of the S-ABCB1 and the S-ABCG2 construct into
pVL1392 vector
For Sf9 cell line transfection, the S-ABCB1 and the S-ABCG2 fragment, respectively, were each
subcloned into the pVL1392 plasmid in its multiple cloning site with the XbaI and NotI restriction sites
(Figure 6.6) giving the pVL1392/S-ABCB1 and pVL1392/S-ABCG2 plasmid, respectively.
Figure 6.6: Vector map of pVL1392 vector (adopted from manual
provided by BD biosciences).
Digestion of ABC transporter fragments and the pVL1392 transfer vector was prepared as described
in section 6.2.3.2. The corresponding agarose gel electrophoresis is illustrated in Figure 6.7.
10000
9632
3902
4000
3000
2000
2027
a
b
c
Figure 6.7: Agarose gel (1%) electrophoresis of digestion
products: pVL1392 vector (a), S-ABCB1 fragment (b),
S-ABCG2 fragment (c).
135
After gel electrophoresis and gel extraction (cf. sections 6.2.3.3 and 6.2.3.4), ligation was performed
at 16 °C overnight in a 20 µL reaction mixture containing 1 µL T4 DNA Ligase and 2 µL of T4 DNA
Ligase Reaction Buffer (10x) as well as the pVL1392 transfer vector (50 ng) and the insert in a molar
ratio of 1:3 (vector to insert). The next day the samples were stored at -20 °C. After transformation of
competent E. coli, Mini-Preps were prepared with colonies picked from ampicillin-containing agar
plates, and a positive clone was chosen for Maxi-Prep.
Plasmid DNA was sequenced using the standard primers pVL1392for and pVL1392rev.
6.2.7
Sf9 insect cell culture and generation of recombinant
baculoviruses
Sf9 insect cell culture and generation of recombinant baculoviruses has been described recently
[Brunskole, 2011]. Sf9 cells (ATCC® CRL-1711™) were cultured in 250 or 500 mL disposable
Erlenmeyer flasks at 28 °C under rotation at 150 rpm in Insect Xpress medium supplemented with
5% (v/v) fetal calf serum. Recombinant baculoviruses encoding the S-ABCG2 construct were
generated in Sf9 insect cells using the BaculoGold™ transfection kit (BD Biosciences, Heidelberg,
Germany) according to manufacturer’s protocol.
After initial transduction, high-titer virus stocks were generated by two sequential virus
amplifications: in the first amplification, 50 mL cell suspension (density: 2.0 x 106 cells/mL) were
infected with 1 mL of the supernatant fluid from the initial transduction. Cells were cultured for one
week in a 125 mL Erlenmeyer flask, resulting in death of the cell population. As a control for
infectious viruses, 3 mL of this culture were transferred into a 25 cm2 flask and kept at 28 °C in
parallel. The supernatant was harvested and stored under light protection at 4 °C.
In a second amplification, cells were cultured at a density of 3.0 x 106 cells/mL (total
volume: 100 mL), infected with 5 mL of the supernatant fluid from the first amplification and
cultured for two days. After 48 hours the supernatant was harvested. Cells showed signs of infection
(altered morphology, viral inclusion bodies), but the majority of cells was still intact. The supernatant
fluid from the second amplification step was stored under protection from light at 4 °C and used as
high-titer virus stock to infect Sf9 cells. All supernatants were routinely filtered through Millex® filters
(0.22 µm pore size; Merck Millipore, Darmstadt, Germany) to avoid bacterial or fungal
contamination.
136
6.2.8
Chapter 6
Recombinant transporter expression in pVL1392/S-ABCG2infected Sf9 cells
6.2.8.1 Immunological detection of ABCG2 expression
To ensure an optimal accumulation of the desired ABCG2 protein, an ideal Sf9 infection time has to
be determined. Cells were seeded at a density of 0.75 x 106 cells/mL, infected with high-titer
baculovirus stock (1:500) and cultured. Cell aliquots were harvested at different times after infection
(24-96 h) followed by three washing and centrifugation steps with PBS at 4 °C for 10 min (3,000 g)
using a Minifuge 1.0R (Heraeus Instruments GmbH; Hanau, Germany). To monitor the expression of
ABCG2 in pVL1392/S-ABCG2-infected Sf9 cells, total cell lysates were analyzed by immunoblotting.
6.2.8.1.1 Sf9 cell lysis
Cell lysis of ABCG2-infected Sf9 cells was performed by homogenizing the harvested cells in RIPA
buffer. After incubation for 20 min on ice in an Eppendorf reaction vessel, the cell suspension was
centrifuged (10 min, 4° C, 16,000 g) and the supernatant containing the proteins was frozen at -80 °C
after addition of 10% (v/v) glycerol.
6.2.8.1.2 SDS-Page and Western Blotting
SDS-PAGE and WB analysis as well as the determination of the protein content of all cell lysates were
performed as described in section 5.2.9. The primary rat monoclonal antibody ABCG2 (BXP-53) was
used in a 1:1,000 dilution, the secondary AB (goat anti-rat IgG-HRP) in a 1:10,000 dilution.
For molecular weight control, the prestained protein marker peqGOLD Protein Marker III
(20-122 kDa; Peqlab, Erlangen, Germany) or the Biotinylated Protein Ladder (9-200 kDa) from Cell
Signaling Technology (Danvers, MA, USA) were used (secondary Anti-biotin, HRP-linked AB 1:1,000),
respectively.
Western Blot analysis was similarly performed for ABCG2 membrane preparations.
6.2.8.2 Membrane preparation
For membrane preparation, 200 mL cultures with 3.0 x 106 cells/mL were infected 1:500 with
S-ABCG2 high-titer virus stock, cultured and harvested 48 hours after infection by centrifugation at
4 °C (170 g, 10 min). The cell pellet was resuspended in 100 mL Tris-mannitol buffer (50 mM Tris,
300 mM mannitol, 0.5 mM phenylmethylsulfonyl fluoride (PMSF); pH 7) and again centrifuged.
137
For membrane preparation the cell pellet was lysed and homogenized (25 strokes) in TMEP (50 mM
Tris, 50 mM mannitol, 1 mM EDTA, 10 µg/mL leupeptin, 10 µg/mL benzamidine, 0.5 mM PMSF, 2 mM
DTT; pH 7) by means of a glass-Teflon tissue homogenizer. Buffers were used as described elsewhere
[Sarkadi et al., 1992] with slight modifications. The use of protease inhibitors is recommended
especially in insect cells, as the yields of recombinant proteins can significantly decrease due to high
proteolytic activity [Gotoh et al., 2001].
After centrifugation for 10 min at 500 g to get rid of nuclei and unbroken cells, the supernatant,
containing the membranes, was carefully removed and centrifuged at 100,000 g for 1 h in a Kontron
TGA-45 ultracentrifuge (Kontron Instruments, Munich, Germany). The spun down membranes were
resuspended in TMEP and homogenized by a 20 mL syringe in 15 strokes. Protein quantification was
performed by the method of Bradford in a Bio-Rad Protein Assay (Bio-Rad Laboratories, Munich,
Germany) according to the manual. Membrane aliquots at a protein concentration of 1.5-2.0 mg/mL
were stored at -80 °C until use. All steps during membrane preparation were performed at 4 °C.
Please confer to section 6.3.3.3.2 for a second modified membrane preparation protocol.
6.2.9
ATPase assay for the human BCRP
6.2.9.1 Principle
Determining the ATPase activity in membranes of Sf9 insect cells expressing the ABCG2 transporter
may provide insight into the mechanism of action of modulators. The ATPase activity correlates with
the liberation of inorganic phosphate (Pi), resulting from ATP hydrolysis when a substrate is
transported. Inorganic Pi reacts with ammonium molybdate to form a complex, in the presence of
the reducing agent ascorbic acid producing an intensively blue colored heteropoly complex
(molybdenum blue), which can then be determined spectrophotometrically (Figure 6.8). The
obtained absorbance is proportional to ATPase activity.
138
Chapter 6
S
S
extracellular
TMD
TMD
NBD
NBD
S
intracellular
ATP
ATP
S
ADP
ADP
Pi
Pi
Figure 6.8: Principle of an ATPase assay: Substrate binding and subsequent ATP hydrolysis results in the release of ADP and
inorganic phosphate. Pi forms a complex with ammonium molybdate (gray sickles) which results in an intensive blue color
after reduction; TMD = transmembrane domain, NBD = nucleotide-binding domain, S = substrate; ATP = adenosine
triphosphate; ADP = adenosine diphosphate; Pi = inorganic phosphate.
6.2.9.2 ATPase assay protocol
The ATPase assay protocol was established in the 96-well plate format according to the manual
provided
by
GenoMembrane
(http://www.genomembrane.com/ATPase_assay_Ver.6.5.n.pdf;
Yokohama, Japan) based on the work of Sarkadi [Sarkadi et al., 1992].
6.2.9.2.1 Reagents and buffers

MOPS-Tris buffer [100 mM] contained 20.93 g/L 3-morpholinopropane-1-sulfonic acid. The
pH of 7.0 was adjusted with 1.7 M Tris (37 °C).

Reaction buffer was made of 50 mM MOPS-Tris, 50 mM KCl, 5 mM sodium azide, 2 mM DTT
(AppliChem; Darmstadt, Germany), 1 mM ouabain and 2 mM EGTA. Aliquots were stored at
-20 °C.

For a 100 mM orthovanadate solution, 0.368 g of sodium orthovanadate were dissolved in
16 mL millipore water, the pH was adjusted to 10 with 1 N HCl or 1 N NaOH and the solution
was heated (100 °C) for 10 min. When cooled down, the solution was filled up to 20 mL.
Reaction buffer was used to prepare a 12 mM working solution (also cf. section 6.3.3.1).

200 mM MgATP solution was made by dissolving 2.21 g of Na2ATP (Boehringer; Mannheim,
Germany) in 10 mL MgCl2 solution [400 mM]. The pH was adjusted to 7.0 with
139
1.7 M Tris (37 °C) and the solution was filled up to 20 mL with millipore water. Aliquots of
1 mL were stored at -20 °C. A 12 mM working solution was made in reaction buffer.

For the indicator solution (always freshly prepared), one part of a solution containing 15 mM
zinc acetate and 35 mM ammonium molybdate were mixed with 4 parts of a 5% (m/v)
ascorbic acid solution (pH 5.0 with 10 N NaOH)

Phosphate standards (0, 0.05, 0.1, 0.25, 0.5, 1.0, 1.5 and 2.0 mM) were obtained by diluting a
1 M NaH2PO4 stock solution with reaction buffer. Small portions were stored at -20 °C.

Reactions were stopped with a 10% (m/v) SDS solution (Sigma).

Test compounds (in DMSO) were dissolved in reaction buffer to 3-fold concentrated stock
solutions.
MgATP (instead of Na2ATP) is essential for the assay system as ABCG2 drug transport requires the
presence of magnesium ions [Litman et al., 2000; Özvegy et al., 2001]. It is furthermore linked to a
drug-stimulated, vanadate-sensitive ATPase activity [Sarkadi et al., 1992; Telbisz et al., 2007].
Vanadate, an analog of inorganic phosphate, is a known inhibitor of many ATPases, including ATPase
activities of ABC transporter [Senior et al., 1995; Chen et al., 2001]. Vanadate-sensitive ATPase
activity (resulting from ABC transporter activity) can thereby be calculated by subtracting vanadateinsensitive from total ATPase activity.
6.2.9.2.2 Instrumental set-up
The ATPase assay for the human ABCG2 transporter was established in the 96-well plate format with
absorbance detection at the GENios Pro microplate reader (TECAN Deutschland GmbH; Crailsheim,
Germany). The optimal detection wavelength of the formed phosphomolybdate complex had to be
determined. Therefore, 5 mL of indicator solution were mixed with 50 mg NaH2PO4. After 20 min the
absorbance spectrum of the obtained colored complex was detected at a Cary 100 UV-Vis
spectrophotometer (Varian; Darmstadt, Germany).
Additionally, the linearity of the absorbance of the phosphate standards after complexation and
reduction was tested prior to any experiments at the above determined optimal wavelength. 60 µL
per standard were incubated with 30 µL of 10% SDS and 200 µL coloring solution for 20 min under
shaking (400 rpm) at 37 °C at a Grant-bio PHMP Thermoshaker for microplates (Grant Instruments;
Cambrigdeshire, UK) and absorbance was measured at the microplate reader.
140
Chapter 6
6.2.9.2.3 Assay procedure
The activity of the ABCG2 transporter is measured as vanadate-sensitive ATPase activity in the
presence of inhibitors or stimulating agents (substrates) in membranes of ABCG2 expressing Sf9 cells
as described earlier [Sarkadi et al., 1992; Glavinas et al., 2007].
The assay was performed on ice in the 96-well plate format on the basis of the manual provided by
Geno Membrane:
Membranes were thawed on ice and carefully homogenized with a pipette. Phosphate standards are
part of every plate to calculate a calibration plot for phosphate (60 µL/well of each phosphate
standard performed in doublet). In the remaining wells, membranes (10 µL/well) were loaded with
either 20 µL of the 3-fold concentrated test compound solutions or 3% (v/v) DMSO (in reaction
buffer) for the negative and background control (8 wells each), respectively. The background control,
measures the inorganic phosphate derived from the reaction buffer and was immediately stopped
with a 10% (m/v) SDS solution (30 µL/well) prior to ATPase reaction. The negative control indicates
the basal ATPase activity of the ABCG2 transporter in the absence of a substrate.
Four wells of each compound concentration and each negative control were overlaid with 10 µL of
the 12 mM o-vanadate solution, the remaining wells with the corresponding volume of reaction
buffer. The plate was shaken to mix the solutions and incubated for 3 min at 37 °C.
Subsequently, the reaction was started by the addition of 20 µL of 12 mM MgATP (except for
phosphate standards). After an incubation time of 1 h (37 °C) under shaking the reaction was
stopped by adding 30 µL/well of 10% SDS (except for thebackground control). Indicator solution was
added (200 µL/well) and after further 30 min of incubation at 37 °C, absorbance of the obtained
phosphomolybdate complex was determined in each well at 830 nm at the GENios Pro microplate
reader when the temperature of the microplate had cooled down.
Instrument settings were as follows: Measurement mode: absorbance; number of reads: 10; plate
definition file GRE96ft.pdf; time between move and flash: 100 ms. All values were corrected to the
background control and the amount of inorganic phosphate generated in each well was calculated
from the correlation equation obtained from phosphate standards on each plate. Data obtained
under the treatment with orthovanadate were subtracted from the data gained in the absence of
vanadate to obtain phosphate liberation resulting from ABCG2 activity. This amount of phosphate
was divided by the incubation time and the protein content per well (corresponding to the used
membrane
preparation)
to
give
the
desired
vanadate-sensitive
ATPase
activity
[nmol Pi/min/mg protein].
For the investigation of ABCG2 inhibitors, the assay was performed as described with the exception
that ATPase activity was stimulated with topotecan [100 µM]. Furthermore, topotecan served as
141
positive control on each plate and was used to determine the degree of inhibition caused by the
highest compound concentration tested. The final DMSO content did not exceed 1% (v/v).
142
6.3
Chapter 6
The presentation of the results is focused on the ABCG2 protein.
6.3.1
Results of DNA sequencing
Plasmid DNA was sequenced using the standard primers pVL1392for and pVL1392rev. For the
plasmid containing the S-ABCG2 construct with nearly 2000 bp, corresponding to about half of the
length of the ABCB1 transporter, a complete Primer Walking DNA Sequencing was performed by
Entelechon (Bad Abbach, Germany). The result confirmed an error-free ABCG2 sequence when
compared to the NCBI database (Figures 6.9-6.10; data not completely displayed).
Seq_1
57
Seq_2
59
Seq_1
117
Seq_2
117
GCCGGCTCCCATCGTGACCTCCAGCCGCAGCGCCTCCCACGCCGGCCGCCGCGCGAGGGG
|||
|
|| || |||| || ||
||
|||| | |
ATCGGGCGCGGATCAGATCTGCAGCGGCCGCATGAAGA-CGATCATCGCCCTGAGCTAC-
116
AGCGCTCGGGCGCGCCGGGTGTGGTTGGGGGAAGGGGTTGTGCCGCGCGCGGGCTGCGTG
||
|| |
|
--ATCTTCTGCCTGGTATTCGCC-------------------------------------
176
143
116
137
* * *
Seq_1
477
AGATAAAAACTCTCCAGATGTCTTCCAGTAATGTCGAAGTTTTTATCCCAGTGTCACAAG
|||||||||||||||||||||||||||||||||||||||||||
-----------------ATGTCTTCCAGTAATGTCGAAGTTTTTATCCCAGTGTCACAAG
536
Seq_2
138
Seq_1
537
GAAACACCAATGGCTTCCCCGCGACAGCTTCCAATGACCTGAAGGCATTTACTGAAGGAG
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
GAAACACCAATGGCTTCCCCGCGACAGCTTCCAATGACCTGAAGGCATTTACTGAAGGAG
596
Seq_2
181
Seq_1
597
CTGTGTTAAGTTTTCATAACATCTGCTATCGAGTAAAACTGAAGAGTGGCTTTCTACCTT
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CTGTGTTAAGTTTTCATAACATCTGCTATCGAGTAAAACTGAAGAGTGGCTTTCTACCTT
656
Seq_2
241
Seq_1
657
GTCGAAAACCAGTTGAGAAAGAAATATTATCGAATATCAATGGGATCATGAAACCTGGTC
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
GTCGAAAACCAGTTGAGAAAGAAATATTATCGAATATCAATGGGATCATGAAACCTGGTC
716
Seq_2
301
Seq_1
717
TCAACGCCATCCTGGGACCCACAGGTGGAGGCAAATCTTCGTTATTAGATGTCTTAGCTG
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TCAACGCCATCCTGGGACCCACAGGTGGAGGCAAATCTTCGTTATTAGATGTCTTAGCTG
776
Seq_2
361
Seq_1
777
CAAGGAAAGATCCAAGTGGATTATCTGGAGATGTTCTGATAAATGGAGCACCGCGACCTG
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CAAGGAAAGATCCAAGTGGATTATCTGGAGATGTTCTGATAAATGGAGCACCGCGACCTG
836
Seq_2
421
Seq_1
837
CCAATTTCAAATGTAATTCAGGTTACGTGGTACAAGATGATGTTGTGATGGGCACTCTGA
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CCAATTTCAAATGTAATTCAGGTTACGTGGTACAAGATGATGTTGTGATGGGCACTCTGA
896
Seq_2
481
Seq_1
897
CGGTGAGAGAAAACTTACAGTTCTCAGCAGCTCTTCGGCTTGCAACAACTATGACGAATC
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CGGTGAGAGAAAACTTACAGTTCTCAGCAGCTCTTCGGCTTGCAACAACTATGACGAATC
956
Seq_2
541
Seq_1
957
ATGAAAAAAACGAACGGATTAACAGGGTCATTCAAGAGTTAGGTCTGGATAAAGTGGCAG
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ATGAAAAAAACGAACGGATTAACAGGGTCATTCAAGAGTTAGGTCTGGATAAAGTGGCAG
1016
Seq_2
601
Seq_1
1017
ACTCCAAGGTTGGAACTCAGTTTATCCGTGGTGTGTCTGGAGGAGAAAGAAAAAGGACTA
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ACTCCAAGGTTGGAACTCAGTTTATCCGTGGTGTGTCTGGAGGAGAAAGAAAAAGGACTA
1076
Seq_2
661
Seq_1
1077
1136
721
GTATAGGAATGGAGCTTATCACTGATCCTTCCATCTTGTTCTTGGATGAGCCTACAACTG
|||||||||||||||||||||
GTATAGGAATGGAGCTTATCA---------------------------------------
Seq_2
Seq_1
1137
GCTTAGACTCAAGCACAGCAAATGCTGTCCTTTTGCTCCTGAAAAGGATGTCTAAGCAGG
1196
Seq_2
742
------------------------------------------------------------
741
180
240
300
360
420
480
540
600
660
720
741
Figure 6.9: Annealing of the N-terminal domain of the ABCG2 transporter (1965 base pairs) from NCBI database with the
nucleotide sequence, obtained by sequencing (blue; 604 base pairs) using primer pVL1392for. The encoding region of the
ABCG2 transporter from NCBI database is shown in red (start codon is underlined). The signal peptide sequence, obtained
in sequencing is shown in green.
144
Chapter 6
Seq_1
1981
Seq_2
530
Seq_1
2041
Seq_2
517
Seq_1
2101
Seq_2
457
Seq_1
2161
Seq_2
397
Seq_1
2221
Seq_2
337
Seq_1
2281
Seq_2
277
Seq_1
2341
Seq_2
217
Seq_1
2401
Seq_2
157
Seq_1
2461
Seq_2
97
GTTAGGATTGAAGCCAAAGGCAGATGCCTTCTTCGTTATGATGTTTACCCTTATGATGGT
|||||||||||||
-----------------------------------------------CCCTTATGATGGT
2040
GGCTTATTCAGCCAGTTCCATGGCACTGGCCATAGCAGCAGGTCAGAGTGTGGTTTCTGT
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
GGCTTATTCAGCCAGTTCCATGGCACTGGCCATAGCAGCAGGTCAGAGTGTGGTTTCTGT
2100
AGCAACACTTCTCATGACCATCTGTTTTGTGTTTATGATGATTTTTTCAGGTCTGTTGGT
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
AGCAACACTTCTCATGACCATCTGTTTTGTGTTTATGATGATTTTTTCAGGTCTGTTGGT
2160
CAATCTCACAACCATTGCATCTTGGCTGTCATGGCTTCAGTACTTCAGCATTCCACGATA
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CAATCTCACAACCATTGCATCTTGGCTGTCATGGCTTCAGTACTTCAGCATTCCACGATA
2220
TGGATTTACGGCTTTGCAGCATAATGAATTTTTGGGACAAAACTTCTGCCCAGGACTCAA
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TGGATTTACGGCTTTGCAGCATAATGAATTTTTGGGACAAAACTTCTGCCCAGGACTCAA
2280
TGCAACAGGAAACAATCCTTGTAACTATGCAACATGTACTGGCGAAGAATATTTGGTAAA
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TGCAACAGGAAACAATCCTTGTAACTATGCAACATGTACTGGCGAAGAATATTTGGTAAA
2340
GCAGGGCATCGATCTCTCACCCTGGGGCTTGTGGAAGAATCACGTGGCCTTGGCTTGTAT
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
GCAGGGCATCGATCTCTCACCCTGGGGCTTGTGGAAGAATCACGTGGCCTTGGCTTGTAT
2400
GATTGTTATTTTCCTCACAATTGCCTACCTGAAATTGTTATTTCTTAAAAAATATTCTTA
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
GATTGTTATTTTCCTCACAATTGCCTACCTGAAATTGTTATTTCTTAAAAAATATTCTTA
2460
AATTTCCCCTTAATTCAGTATGATTTATCCTCACATAAAAAAGAAGCACTTTGATTGAAG
|
A-----------------------------------------------------------
2520
518
458
398
338
278
218
158
98
97
Figure 6.10: Annealing of the C-terminal domain of the ABCG2 transporter from NCBI database with the nucleotide
sequence, obtained by sequencing (blue; anti-parallel; 434 base pairs) using primer pVL1392rev. The encoding region of the
ABCG2 transporter from NCBI database is shown in red (stop codon is underlined).
6.3.2
ABCG2 expression in Sf9 cells
6.3.2.1 Infection time
The timing of harvesting the Sf9 cells after ABCG2-baculovirus infection seems to be critical. As
shown in a recently published work, size and morphology of Sf9 cells change dramatically during
infection: The membrane integrity was disrupted at later stages (>72 h) as numerous pores were
observed by means of scanning electron microscopy [Saito et al., 2006; Saito et al., 2009]. Membrane
vesicles prepared from such porous cells are useless. Western blot analysis (total cell lysate) of
pVL1392/S-ABCG2-infected Sf9 cells revealed an optimal transporter expression after 48 h
(Figure 6.11).
145
140
80
50
40
20
a
b
c
d
e
f
g
h
i
Figure 6.11: Expression of the ABCG2 gene product in Sf9 cells. Western blot analysis of total cell lysates prepared from Sf9
wt cells (a) and pVL1392/S-ABCG2-infected Sf9 cells at different times after infection: 24 h (b), 48 h (c), 72 h (d), 96 h (e);
10 µg protein were loaded on lanes a-e; lanes f-h correspond to lanes b-d except for loading with 20 µg protein; i)
Biotinylated Protein Ladder (9-200 kDa; Cell Signaling); the sizes of molecular weight standards are indicated in kDa.
Closer examination of the Western blot with the Quantity One (Bio-Rad) software unfortunately
revealed a main band (presumably a doublet) at 63-67 kDa (expected molecular weight of the
glycosylated ABCG2 transporter monomer: 72 kDa). Additionally, some diffuse bands were visible
probably resulting from proteolytic cleavage of the target ABCG2 protein into lower molecular
weight products when expressed in Sf9 cells.
To confirm this lower molecular mass of the ABCG2 transporter in insect cells, the expression of BCRP
in membrane preparations was compared to that in MCF-7/Topo cells (ABCG2 overexpressing
subclone of human MCF-7 cells). Immunoblot analysis (Figure 6.12) showed that ABCG2 expressed in
Sf9 cells migrates as a double band of lower apparent molecular mass (63 kDa and 66 kDa) than the
ABCG2 protein of MCF-7/Topo cells (a wide band at 74 kDa). This phenomenon was previously
reported by other groups [Özvegy et al., 2001; Pozza et al., 2006; Pozza et al., 2009].
a
b
c
d
140
80
Figure 6.12: Immunoblot detection of the ABCG2 transporter
expressed in Sf9 membrane preparations and MCF-7/Topo cells,
respectively (10 µg of protein); a) Sf9 wt membrane, b) Sf9
ABCG2-membrane, c) Biotinylated Protein Ladder (9-200 kDa;
Cell Signaling), d) MCF-7/Topo cells. Red arrow indicates the
ABCG2 transporter (72 kDa) expressed in human MCF-7/Topo
cells. Immunoblot analysis revealed an approximately 5-fold
higher expression level of the lower ABCG2 homolog (doublet
protein bands) compared to the ABCG2 monomer in
MCF-7/Topo cells. BXP-53 (1:1,000) was used as primary AB.
50
40
a
b
c
d
146
Chapter 6
The high extent of proteolysis of the ABCG2 transporter in Sf9 cells is remarkable. To check whether
the formation of the ABCG2 homolog at lower molecular mass and the proteolytic effect in Sf9 cells
depends on the time of infection and the dilution of the virus stocks, Sf9 membane preparations
were made 24, 48 and 72 h after infection, each charged 1:100 and 1:500 with S-ABCG2 high titer
virus stock, respectively (Figure 6.13).
a b
c d e
f
g h
i
j
140
80
50
40
A
116
66
45
35
B
Figure 6.13: A) Immunoblot analysis of ABCG2
protein expression in Sf9 membrane preparations
and MCF-7/Topo cells, respectively; B) corresponding
Coomassie blue-stained gel.
Cells were harvested at various times after infection
(lanes d-i) with different dilutions of ABCG2 high titer
virus stock (stated in brackets): a) peqGold protein
marker I (14-116 kDa; Peqlab); b) MCF-7/Topo cells;
c) Sf9 wt membrane; Sf9 ABCG2 membrane: d) 24 h
(1:500), e) 24 h (1:100), f) 48 h (1:500), g) 48 h
(1:100), h) 72 h (1:500), i) 72 h (1:100); j) Biotinylated
Protein Ladder (9-200 kDa; Cell Signaling). All
samples (10 µg of total protein) were subjected to
SDS-PAGE on 12% polyacrylamide gels followed by
immunoblotting with the rat monoclonal ABCG2
antibody BXP-53 (1:1,000).
The ratio of proteolytic degradation products (bands around 50 kDa) compared to the target ABCG2
protein at 63 kDa is nearly constant in the Sf9 membrane preparations, virtually irrespective of the
infection time and the used dilution of the virus stock. However, membranes prepared 24 h after
infection of the cell showed the lowest extent of proteolysis. Despite a lower expression level of the
ABCG2 homolog in comparison to 48 h- and 72 h-preparations, these short-term infected
membranes were preferred for the first ATPase experiments.
6.3.2.2 Glycosylation of the ABCG2 transporter
The ABCG2 transporter is a glycoprotein consisting of 655 amino acids with a predicted molecular
weight of 72 kDa. As a ‘half-transporter’ it contains only one TMD forming a homodimer in the
plasma membrane in order to function. The topology of human ABCG2 is displayed in Figure 6.14.
147
596
600
557
418
500
extracellular
OUT
membrane
MEMBRANE
400
IN
intracellular
ATP
SITE
NBD
C Terminus
C-terminus
200
100
MXR SNPs:
V12M
Q141K
E366E
L475L
D620N
300
N Terminus
N-terminus
Figure 6.14: Topology model of the human ABCG2 half-size transporter consisting of 6 transmembrane domains and one
nucleotide binding domain (NBD). Potential N-linked glycosylation sites at amino acid positions 418, 557 and 596 are
displayed in blue (modified from [Ejendal and Hrycyna, 2002; Diop and Hrycyna, 2005]).
Diop and Hrycyna [Diop and Hrycyna, 2005] showed that stably expressed ABCG2 in membranes
from human cells was only glycosylated at the N596 site, and treatment with endoglycosidase
PNGase F (peptide-N-glycosidase F) reduced the molecular mass on SDS-PAGE gels to approximately
60 kDa. The phenomenon of a modified molecular weight due to deglycosylation has also been
observed for ABCG2 expressed in MCF-7 cells upon treatment with PNGase F or the nucleoside
antibiotic tunicamycin [Litman et al., 2002] as well as in Sf9 insect cells [Özvegy et al., 2001; Pozza et
al., 2009; Jacobs, 2010] and yeast systems [Mao et al., 2004]. In mammalian systems, the ABCG2
transport protein is N-glycosylated. Yeasts and insect cells show a reduced glycosylation potential
[Harrison and Jarvis, 2006] and the original human protein is therefore ‘underglycosylated’.
It was reported that the disparity in the extent of glycosylation did not affect ATPase activity [Diop
and Hrycyna, 2005; Pal et al., 2007] and the ABCG2 transporter was functionally active in ATPase
assays using membrane preparations from Sf9 cells [Özvegy et al., 2001; Glavinas et al., 2007; Pozza
et al., 2009].
148
Chapter 6
6.3.3
Optimization, validation and application of the ABCG2
ATPase assay
6.3.3.1 Set-up of the ATPase assay for the human BCRP
Prior to any experiments, the instrumental set-up of the planned ATPase assay in the 96-well format
had to be elucidated. Therefore, the optimal detection wavelength of the obtained blue colored
phosphomolybdate complex (after reduction with ascorbic acid) was identified (Figure 6.15).
Considering the spectral intensity of the arc-discharge xenon lamp of the plate reader, the detection
wavelength was set to 830 nm.
In the concentration range from 0.05 to 2 mM the used NaH2PO4 standards show excellent linearity
regarding absorbance at 830 nm (Interference filter 830-15/75; Eureca Messtechnik GmbH, Köln,
Germany) after complexing and were therefore appropriate for the assay.
3.0
1.8
1.6
B
A
2.5
580
1.2
Absorbance A
Absorbance
1.4
1.0
0.8
0.6
0.4
2.0
1.5
1.0
0.5
0.2
0.0
400
R2 = 0.998
0.0
500
600
700
wavelength [nm]
800
900
0
20
40
60
80
100
120
phosphate [nmol]
Figure 6.15: A) Absorbance spectrum of the reduced phosphomolybdate complex recorded with a Cary 100 UV-Vis
spectrophotometer; B) Absorbance of the phosphate standards after complexation and reduction measured at the GENios
Pro microplate reader at 830 nm.
Membrane preparations from Sf9 cells after an infection time of 24 h showed an ATPase activity of
about 25 nmol Pi/min/mg protein when treated with 100 µM topotecan and prazosin, respectively.
As shown in Figure 6.16, sodium orthovanadate at a concentration of 5 mM nearly completely
inhibited vanadate-sensitive ATPase activity. The final concentration in the assay was set to 2 mM
due to an unspecific color reaction of vanadate with the indicator solution at higher concentrations.
149
ATPase activity
[nmol Pi/min/mg protein]
25
Figure 6.16: ATPase activity of ABCG2-membranes
(Infection: 24 h, 1:500) in the presence of different
concentrations of sodium orthovanadate was reduced
from 25 to 7 nmol Pi/min/mg protein (corresponding to
vanadate-insensitive ATPase activity) when stimulated
with topotecan [100 µM]. Data were obtained from two
independent experiments.
20
15
10
5
0
0
1
2
3
4
5
6
7
8
9
10 11
concentration [mM]
First experiments on the ABCG2 ATPase activity, performed in the presence of the transporter
substrates topotecan, mitoxantrone (Figure 6.17) or prazosin, using membrane preparations from
Sf9 cells infected for 24 h, revealed a very high basal membrane ATPase activity. Under these
conditions the sensitivity of
the
ATPase
assay
0.30
was
0.25
effects of the stimulatory
drug relative to the negative
Absorbance
substantially reduced, as the
0.20
0.15
0.10
control (2% (v/v) DMSO)
were quite low. By reducing
the final DMSO amount to a
maximum
of
1%,
the
0.05
0.00
450 500 550 600 650 700 750 800 850
wavelength [nm]
Figure 6.17:
Absorbance spectrum
of the ABCG2 substrate
mitoxantrone
with
maxima at 612 and
664 nm. Mitoxantrone
gives a dark-blue color
which
does
not
interfere
with
the
detection
of
the
molybdate complex in
the ATPase assay at
830 nm.
difference between the negative (DMSO) control and topotecan could, indeed, be increased but was
still insufficient (data not shown).
The molecular background of this high basal activity was extensively studied but is not known till
now. It is speculated that the ABCG2 transporter is constitutively active, that is, the high basal activity
may result from uncoupled (transport-independent) ATPase activity. The presence of unknown
substrates in the membrane compartment [Sarkadi and Telbisz, 2013] and ineffective
homodimerization of ABCG2 to a ‘full transporter’ have to be taken into account as well.
The problem of low sensitivity and a high basal activity of ATPase assays for the ABCG2 transporter
has been described in literature [Özvegy et al., 2001; Özvegy et al., 2002; Mao et al., 2004] and
different experiments have been conducted to solve the problem by preparing ABCG2 membranes
with a fully active transporter protein to guarantee a high-sensitivity screening assay [Pal et al., 2007;
Telbisz et al., 2007].
150
Chapter 6
To exclude that the insensitivity of the assay has to be attributed to an insufficient expression level of
the ABCG2 protein after 24 h of infection, additional experiments with membranes of cells with a
longer infection time were performed in the presence of the ABCG2 substrates topotecan, prazosin
and mitoxantrone and potent BCRP inhibitors including fumitremorgin C (FTC), Ko143 and different
synthesized modulators (cf. Chapter 3). In fact, with ABCG2 overexpressing membranes of Sf9 cells
with an infection time of 48 h, the difference between basal and substrate stimulated activity could
be improved, but was still inadequate (Figure 6.18, bars 1-4).
6.3.3.2 Mode of ATPase inhibition
There was a significant difference in vanadate-sensitive ATPase activity between substrate-treated
and inhibitor-treated ABCG2 membranes (Figure 6.18). Topotecan-, mitoxantrone- and prazosinstimulated ATPase activities (bars 2-4) were reduced by about 30-40% from approximately 30 to
20 nmol Pi/min/mg protein with the addition of 5 µM fumtremorgin C (bars 6-8). Compound 51
reduced the basal activity to the same extent, indicating that both FTC and 51 act as real ABCG2
vanadate-sensitive ATPase activity
inhibitors, not as substrates.
35
Figure 6.18: Vanadate-sensitive ATPase
activity of ABCG2 membranes (48 h; 1:500)
when treated with ABCG2 substrates,
ABCG2 inhibitors or a combination of both:
1) Basal activity (DMSO 1%);
2) Topotecan [100 µM];
3) Mitoxantrone [5 µM];
4) Prazosin [5 µM];
5) FTC [5 µM];
6) Topotecan [100 µM] + FTC [5 µM];
7) Mitoxantrone [5 µM] + FTC [5 µM];
8) Prazosin [5 µM] + FTC [5 µM];
9) UR-MB108-4 (51) [10 µM].
Activity was calculated from 2-3
independent experiments.
30
25
20
15
10
5
0
1
2
3
4
5
6
7
8
9
A selection of synthesized modulators was investigated for concentration-dependent inhibition.
Data are summarized in Table 6.1. The inhibitory effects obtained in the ATPase assay were not
directly comparable to those in the Hoechst assay: The percental inhibition caused at the highest
concentration tested in the ATPase assay refers to the drug-stimulated activity caused by 100 µM of
topotecan, whereas the maximal inhibition in the Hoechst assay is referred to the inhibition caused
by fumitremorgin C at 10 µM (100% inhibition). Nevertheless, data obtained in the two assay systems
correlate quite well. IC50 values from the two assays are in a good correlation especially for highly
potent compounds such as 3, 32, 51 and the standard compounds FTC and Ko143. Compounds with
low efficiency in the Hoechst assay, for example 22, also show reduced maximal effects in the ATPase
151
assay and vice versa (see compound 32). As an example, the inhibitory effect of FTC and 51 in the
ATPase assay is displayed in Figure 6.19 in a concentration-dependent manner, giving reproducible
sigmoidal concentration response curves. In general, the ATPase assay gave lower IC50 values than
the Hoechst 33342 assay. This was particularly obvious for compounds 22 and 42. Apperently, the
(partial) proteolytic cleavage of the ABCG2 transporter when expressed in Sf9 cells (cf. section
6.3.2.1) does not affect the function of the membranes in the ATPase inhibition mode.
20
18
16
14
12
10
8
6
4
A
0.001
0.01
0.1
1
18
16
14
12
10
8
B
6
10
0.001
concentration [µM]
0.01
0.1
1
concentration [µM]
Figure 6.19: Vanadate-sensitive ATPase activity of ABCG2 membranes under the treatment of the potent BCRP inhibitors
fumitremorgin C (A) and compound 51 (B).
Table 6.1: ABCG2 inhibition by selected compounds in the ATPase and the Hoechst 33342 assay.
ATPase assaya
a
Hoechst 33342 assayb
Compound
IC50 [nM]c
Max. Inhibition
(%)d
IC50 [nM]
Max. Inhibition
(%)e
FTC
603 ± 161
62 ± 10
731 ± 92
100
Ko143
73 ± 14
71 ± 8
117 ± 5f
103 ± 7f
3 (UR-COP77)
65 ± 18
60 ± 8
183 ± 32g
83 ± 5g
7 (UR-COP145)
143 ± 40
57 ± 7
943 ± 79
87 ± 5
12 (UR-COP228)
178 ± 28
63 ± 5
591 ± 87
109 ± 8
22 (UR-COP238)
692 ± 74
35 ± 2
2,238 ± 202
22 ± 0
32 (UR-COP269)
37 ± 14
69 ± 5
59 ± 11
101 ± 5
42 (UR-QQS-2)
176 ± 59
59 ± 6
1,043 ± 53
107 ± 7
51 (UR-MB108-4)
23 ± 11
59 ± 4
64 ± 3
95 ± 3
Data from inhibition of topotecan-stimulated [100 µM] ATPase activity on membranes of Sf9 cell infected for 48 h and
72 h, respectively (cholesterol-free); mean values ± SEM from two to three independent experiments performed in
b
quadruplicate; Hoechst 33342 microplate assay (ABCG2 overexpressing MCF-7/Topo cells); mean values ± SEM from two
c
to three independent experiments performed in sextuplicate; Respective concentration of the inhibitor required for halfd
maximal inhibition; expressed as percental inhibition caused by the highest concentration of the test compound (reduced
e
ATPase activity) relative to 100 µM topotecan (providing full ATPase activity); Maximal inhibitory effects expressed as
percental inhibition caused by the highest concentration of the compound tested relative to the reference compound FTC
f
g
[10 µM] (100% inhibition); [Kühnle, 2010]; [Ochoa-Puentes et al., 2011]
152
Chapter 6
The results revealed that the synthesized modulators inhibited the topotecan-stimulated ATPase
activity, indicating that these compounds are not substrate of the ABCG2 transporters, but inhibitors.
To confirm the mechanism of inhibition and the reliability of the assay, drugs working as inhibitors
and poorly transported substrates should be included in the test series. With regard to investigations
on competitive substrates/inhibitors in the ATPase assay, caution should be taken that an inhibitor of
ABCG2 and its potency is dependent on the used substrate and the respective concentrations.
Although competitive substrates have been identified for the ABCB1 transporter in drug
accumulation studies [Scala et al. 1997], only few information is available about competitive ABCG2
substrates/inhibitors. Lately, imatinib mesylate, was shown to reverse BCRP-mediated resistance to
topotecan and SN-38 (active metabolite of irinotecan) in vitro suggesting that it directly inhibits
BCRP-mediated transport not being a competitive substrate [Houghton et al., 2004]. In contrast to
that, Burger et. al. considered imatinib mesylate as a substrate of ABCG2 [Burger et al., 2004].
Furthermore, the inhibitory effect of the ABCG2 substrate flavopiridol on mitoxantrone efflux in
ABCG2-overexpressing cells in high concentrations was attributed to competitive inhibition
[Robey et al., 2001].
Despite low water solubility, the synthesized compounds worked well in this protein-free buffer
system, although the ABCG2-ATPase activity was not completely inhibited. As the Hoechst 33342
assay was performed in the presence of FCS, the question arose whether albumin supplementation
might increase the maximum effects of the compounds in the ATPase assay. The concentrationresponse curves of the ABCG2 inhibitor 12 on the ATPase activity in the presence and absence of
5% (m/v) BSA is shown in Figure 6.20. The addition of serum albumin did obviously not affect the
ATPase inhibitory effect of 12 in this type of assay.
ATPase inhibition (%)
80
60
Figure 6.20: Effect of 5% (m/v) BSA (triangles) on the
inhibition of ABCG2-ATPase activity by compound 12 in Sf9
membrane preparations; percental inhibition is determined
as ratio of ATPase activity at the respective concentration
referred to the topotecan-stimulated [100 µM] activity in
absence of an inhibitor.
40
20
0
0.0001
0.001
0.01
0.1
concentration [µM]
1
10
153
6.3.3.3 Cholesterol-loaded ABCG2 Sf9 membranes
It has been known for a long time that insects are unable to synthesize sterols de novo and require an
exogenous supply of sterols [Clayton, 1964]. Cholesterol is the precursor of steroid hormones, for
example cortisol in mammals and ecdysone in insects, and an essential constituent of membranes in
animals [Ikekawa et al., 1993]. Cholesterol plays a crucial role in the function of membrane proteins.
Previously, it was demonstrated that the cholesterol levels in insect cells are a limiting factor in the
expression of the oxytocin receptor in the baculovirus/Sf9 system [Gimpl et al., 1995]. With the
addition of cholesterol either to the insect cell culture medium or to membrane preparations, the
specific binding of oxytocin to the oxytocin receptor was significantly increased in a concentrationdependent manner. Cholesterol-supplementation, however, produced controversial results
regarding growth and morphology of insect cells, although lipids were shown to be essential for
baculovirus infection and subsequent protein expression [Gimpl et al., 1995; Gilbert et al., 1996; Lua
and Reid, 2005].
6.3.3.3.1 Effect of cholesterol on ATPase activity of ABCG2-Sf9 membrane preparations
By analogy with the above described results, Pal et al. observed a lack of ABCG2-ATPase activity
stimulation by the BCRP substrates prazosin (Figure 6.22) and topotecan referred to the basal activity
when expressed in Sf9 cell membranes, whereas ABCG2 overexpressing human cell membranes and
cholesterol-loaded Sf9 membranes could be activated. The authors showed that the low cholesterol
content of insect cells decisively influences ABCG2 stimulation, as cholesterol loading of membrane
preparations significantly potentiates the stimulation of basal ABCG2 activity [Pal et al., 2007; Telbisz
et al., 2007]. Therefore, membranes of Sf9 cells (48 h of infection) were thawed and incubated on ice
for 20 min with a complex of randomly methylated-β-cyclodextrin (RAMEB) and cholesterol (final
content: 5.4%) purchased from Cyclolab (Cyclodextrin Research and Development Laboratory;
Budapest, Hungary) at a final concentration of 2.5 mg/mL. The membranes were centrifuged at
16,000 to remove residual cholesterol and, subsequently, resuspended in TMEP buffer (cf. 6.2.8.2).
Wildtype ABCG2 membranes and cholesterol-loaded membranes were treated with ABCG2
substrates topotecan and prazosin, respectively, and the ATPase stimulation was compared. As
shown in Figure 6.21 the subsequent incubation with the RAMEB-cholesterol complex led to a
1.5-fold stimulation of basal activity for topotecan and prazosin, respectively, compared to 1.1 for wt
membranes without cholesterol. The final ATPase stimulation in these membrane preparations is
summarized in Table 6.2.
154
Chapter 6
vanvadate-sensitive ATPase activity
30
25
20
Figure 6.21: Vanadate-sensitive ATPase activity of ABCG2 Sf9 wt
membranes (1) and cholesterol-loaded membranes (2) when
incubated (1 h) with topotecan [100 µM] (black bars), prazosin
[100 µM] (grey bars) and DMSO (=basal activity; white bars);
results were obtained from 3 independent experiments each
performed in quadruplicate.
15
10
5
0
1
2
Addition of cholesterol to ABCG2-Sf9 membranes significantly increased ATPase activity, suggesting
that the low cholesterol content in Sf9 cells severely affected the activation of the ABCG2
transporter. It is difficult to decide whether cholesterol directly increases ABCG2-ATPase activity or
has an indirect effect by altering the architecture and the fluidity of the phospholipid bilayer in the
plasma membrane. Only a few studies addressing the effect of cholesterol on ABC transporters have
been published so far, especially for the ABCB1 transporter [Garrigues et al., 2002; Arima et al.,
2004]. Recently, it has been suggested that cholesterol is a substrate of ABCB1 [Le Goff et al., 2006].
This may also be taken into consideration in case of ABCG2 as other members of the ABCG subfamily
were reported to be involved in cholesterol transport [Pal et al., 2007], e.g. ABCG5 and ABCG8 half
transporters forming heterodimers for intestinal absorption and biliary excretion of cholesterol
[Graf et al., 2003].
6.3.3.3.2 Modified membrane preparation protocol
Based on these encouraging results, the original membrane preparation protocol described in section
6.2.8.2 was slightly modified:
Prior to the last centrifugation step at 100,000 g, membranes were incubated with the RAMEBcholesterol complex at a final concentration of 2.5 mg/mL for 20 min at 4 °C. After resuspension and
protein quantification, membranes were subsequently aliquoted and stored at -80 °C until use.
6.3.3.4 Effect of CHAPS on basal and drug-stimulated ABCG2-ATPase activity
In a recently published patent by Sarkadi and Telbisz [Sarkadi and Telbisz, 2013], the difference
between drug-stimulated and basal activity could be increased in the presence of CHAPS
(3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate),
a
non-denaturing
zwitterionic
sulfobetaine detergent which is commonly used for protein solubilization and disruption of proteinprotein interactions (Figure 6.22).
155
Figure 6.22: Chemical structures of the sulfobetaine-type zwitterionic detergent CHAPS (a) and the ABCG2 substrate
prazosin (b). The flavonoid quercetin (c) was recently shown to be transported by ABCG2 [Sesink et al., 2005].
Cholesterol-loaded membranes (prepared as described in 6.3.3.3.2) were used for the ATPase assay
in the presence and absence of prazosin [100 µM] under simultaneous treatment with different
concentrations of CHAPS [0.1-2.0 mM] (Figure 6.23).
ATPase activity
40
38
36
34
32
30
Figure 6.23: Effect of of CHAPS on the basal (circles) and
prazosin-stimulated (squares) ABCG2-ATPase activity in
cholesterol-loaded Sf9 membranes.
28
26
0.0
0.5
1.0
1.5
2.0
2.5
CHAPS [mM]
With the addition of CHAPS to the cholesterol-loaded membrane preparation the basal ATPase
activity was significantly reduced about 25% whereas the drug-stimulated activity remained nearly
unchanged. Accordingly, the protocol of the ATPase assay was modified: a final concentration of
1 mM CHAPS was added to the membrane preparations immediately prior to the experimental
procedure.
The differences in topotecan-, quercetin- and prazosin-stimulated and FTC-inhibited ABCG2-ATPase
activity using ‘ABCG2 wildtype’ and cholesterol-loaded membranes in the presence or absence of
CHAPS become obvious from Figure 6.24.
156
Chapter 6
40
Figure 6.24:
Vanadate-sensitive ATPase activity of
ABCG2 Sf9 wt (1), cholesterol-loaded (2)
and cholesterol-loaded membranes
modulated with CHAPS [1 mM] (3) when
incubated with topotecan [100 µM]
(black), prazosin [100 µM] (red),
quercetin
[10
µM]
(blue),
fumitremorgin C [10 µM] (gray) and
DMSO
(=basal
activity;
white),
respectively; n = 3.
30
20
10
0
1
2
3
Table 6.2: Effect of cholesterol and CHAPS on the stimulation or inhibition of the ABCG2-ATPase in
ABCG2 Sf9 membranes in presence of various ABCG2 modulating agents. Stimulation is expressed as
the ratio of the drug-stimulated and the basal ATPase activity, inhibition is given as the percental
reduction of the basal activity.
wt ABCG2
membranes
cholesterolloaded ABCG2
membranes
cholesterol-loaded
ABCG2 membranes +
CHAPS [1 mM]
Compound
concentration
[µM]
fold stimulation
fold stimulation
fold stimulation
Topotecan
100
1.32
1.51
1.77
Prazosin
100
1.25
1.43
1.90
Quercetin
10
2.47
2.36
2.94
Inhibition
Inhibition
Inhibition
25.4%
35.7%
7.2%
Fumitremorgin C
10
In summary, with the addition of cholesterol to the membrane preparation the ratio of stimulated
and basal ATPase activity was significantly improved. The treatment with the modulation agent
CHAPS led to an additional enhancement of sensitivity in the stimulation mode of the assay by
reducing the basal ATPase activity. FTC was not able to further lower the reduced basal activity in
presence of CHAPS suggesting that the minimum of basal activity is reached with the addition of the
detergent.
Figure 6.25 shows the activation of basal ATPase activity by ABCG2 substrates topotecan, prazosin
and quercetin in cholesterol-loaded (+ 1 mM CHAPS) and wt ABCG2 membranes in a concentrationdependent manner.
22
A
vanadate-sensitive ATPase acticity
vanadate-sensitive ATPase acticity
20
18
16
14
12
22
B
20
18
16
14
12
10
8
10
0.1
1
10
0.1
100
1
30
10
100
concentration [µM]
concentration [µM]
157
C
25
Figure 6.25: ABCG2-ATPase activation of ABCG2 wt
(circles) and cholesterol-CHAPS-loaded Sf9 membranes
(squares) upon topotecan (A), prazosin (B) and
quercetin (C) treatment.
20
15
10
0.001
0.01
0.1
1
10
concentration [µM]
The comparison of ABCG2 wt and cholesterol-loaded membranes (without CHAPS) in the inhibition
mode is exemplarily shown for compounds 32 and 52 in Figure 6.26. The use of cholesterol generally
resulted in higher ATPase activities upon stimulation by topotecan, but did not influence potency and
efficacy parameters to a noteworthy extent (Table 6.3).
vanadate sensitive ATPase activity
25
25
20
15
10
5
A
0.0001
0.001
0.01
concentration [µM]
0.1
1
20
15
10
5
B
0
0.001
0.01
0.1
1
concentration [µM]
Figure 6.26: Inhibition of ABCG2-ATPase by compound 32 (A) and 52 (B) after stimulation with topotecan [100 µM] in
ABCG2 wt (circles) and cholesterol-loaded Sf9 membranes (squares).
In general, cholesterol-loaded membranes allowed for the construction of better reproducible
sigmoidal concentration-response curves with higher amplitude, resulting in lower errors of the IC50
values.
158
Chapter 6
Table 6.3: Inhibitory activities of selected ABCG2 modulators, determined in the ATPase assay on
ABCG2 wt and cholesterol-loaded membranes a
wt ABCG2 membranes
a
cholesterol-loaded ABCG2 membranes
Compound
IC50 [nM]
Imax (%)
IC50 [nM]
Imax (%)
32
32 ± 11
60 ± 8
22 ± 3
65 ± 5
34
75 ± 13
79 ± 7
46 ± 12
75 ± 7
52
47 ± 25
79 ± 10
45 ± 11
85 ± 5
Mean values ± SEM calculated from two independent experiments performed in quadruplicate.
6.4
The ABCG2 template in combination with the signal peptide sequence for expression in Sf9 insect
cells was successfully and error-free cloned into the pVL1392 vector. Recombinant baculoviruses
encoding the ABCG2 transporter were generated and used for Sf9 cell infection. To achieve ABCG2
overexpression in Sf9 cell membranes, cells were infected for 48 h with high titer virus stocks despite
a higher extent of proteolysis of the target ABCG2 protein when compared to cells infected for a
period of 24 h.
In insect cells, the ABCG2 monomer was expressed at a lower molecular weight due to
deglycosylation. However, the difference in the extent of glycosylation did not affect the ATPase
activity of the transporter.
Very high basal activity severely compromised assay analysis. The low sensitivity caused high errors
and an inacceptable signal-to-noise ratio. The narrow margin between basal and drug-stimulated
ATPase activity of ABCG2 Sf9 membranes was significantly improved by incubating ABCG2membranes with a cyclodextrin-cholesterol complex. The concomitant addition of CHAPS to
cholesterol-loaded membrane preparations decreased the basal activity and led to a sufficient
difference between basal and stimulated ABCG2-ATPase activity to establish an assay protocol. The
ATPase inhibition mode produced slightly higher maximal effects when cholesterol was added to the
membranes, but worked likewise with wildtype ABCG2 Sf9 membranes.
The results obtained from the ATPase assay demonstrated that the synthesized ABCG2 modulators
described in Chapter 3 are inhibitors and not substrates. The established ATPase assay for the human
ABCG2 transporter on Sf9 membrane preparations proved to be well-suited system to discriminate
between ABCG2 substrates and inhibitors. In combination with other assays, the determination of
ATPase activity harbors the potential of being used for the profiling of ABCG2 modulators to detect
different mechanisms of drug-transporter interaction.
6.5
159
References
Arima, H., Yunomae, K., Morikawa, T., et al. Contribution of cholesterol and phospholipids to
inhibitory effect of dimethyl-beta-cyclodextrin on efflux function of P-glycoprotein and
multidrug resistance-associated protein 2 in vinblastine-resistant Caco-2 cell monolayers.
Pharm. Res. 2004, 21(4), 625-634.
Brunskole, I. Molecular and Cellular Analysis of Aminergic G Protein-Coupled Receptors: Histamine
H2, H4 and β2-Adrenergic Receptors, a Scientific Paradigm. PhD Thesis, University of
Burger, H., van Tol, H., Boersma, A. W., et al. Imatinib mesylate (STI571) is a substrate for the breast
cancer resistance protein (BCRP)/ABCG2 drug pump. Blood 2004, 104(9), 2940-2942.
Chen, J., Sharma, S., Quiocho, F. A., et al. Trapping the transition state of an ATP-binding cassette
transporter: evidence for a concerted mechanism of maltose transport. Proc. Natl. Acad. Sci.
U S A 2001, 98(4), 1525-1530.
Clayton, R. B. The Utilization of Sterols by Insects. J. Lipid Res. 1964, 5, 3-19.
Diop, N. K. and Hrycyna, C. A. N-linked glycosylation of the human ABC transporter ABCG2 on
asparagine 596 is not essential for expression, transport activity, or trafficking to the plasma
membrane. Biochemistry 2005, 44(14), 5420-5429.
Ejendal, K. F. and Hrycyna, C. A. Multidrug resistance and cancer: the role of the human ABC
transporter ABCG2. Curr. Protein Pept. Sci. 2002, 3(5), 503-511.
Garrigues, A., Escargueil, A. E. and Orlowski, S. The multidrug transporter, P-glycoprotein, actively
mediates cholesterol redistribution in the cell membrane. Proc. Natl. Acad. Sci. U S A 2002,
99(16), 10347-10352.
Germann, U. A., Willingham, M. C., Pastan, I., et al. Expression of the human multidrug transporter in
insect cells by a recombinant baculovirus. Biochemistry 1990, 29(9), 2295-2303.
Gilbert, R. S., Nagano, Y., Yokota, T., et al. Effect of lipids on insect cell growth and expression of
recombinant proteins in serum-free medium. Cytotechnology 1996, 22(1-3), 211-216.
Gimpl, G., Klein, U., Reilander, H., et al. Expression of the human oxytocin receptor in baculovirusinfected insect cells: high-affinity binding is induced by a cholesterol-cyclodextrin complex.
Biochemistry 1995, 34(42), 13794-13801.
Glavinas, H., Kis, E., Pal, A., et al. ABCG2 (breast cancer resistance protein/mitoxantrone resistanceassociated protein) ATPase assay: a useful tool to detect drug-transporter interactions. Drug.
Metab. Dispos. 2007, 35(9), 1533-1542.
Gotoh, T., Miyazaki, Y., Sato, W., et al. Proteolytic activity and recombinant protein production in
virus-infected Sf-9 insect cell cultures supplemented with carboxyl and cysteine protease
inhibitors. J. Biosci. Bioeng. 2001, 92(3), 248-255.
Graf, G. A., Yu, L., Li, W. P., et al. ABCG5 and ABCG8 are obligate heterodimers for protein trafficking
and biliary cholesterol excretion. J. Biol. Chem. 2003, 278(48), 48275-48282.
160
Chapter 6
Harrison, R. L. and Jarvis, D. L. Protein N-glycosylation in the baculovirus-insect cell expression system
and engineering of insect cells to produce "mammalianized" recombinant glycoproteins. Adv.
Virus Res. 2006, 68, 159-191.
Houghton, P. J., Germain, G. S., Harwood, F. C., et al. Imatinib mesylate is a potent inhibitor of the
ABCG2 (BCRP) transporter and reverses resistance to topotecan and SN-38 in vitro. Cancer
Res. 2004, 64(7), 2333-2337.
Ikekawa, N., Morisaki, M. and Fujimoto, Y. Sterol-Metabolism in Insects - Dealkylation of Phytosterol
to Cholesterol. Accounts Chem. Res. 1993, 26(4), 139-146.
Jacobs, A. Expression, Aufreinigung und funktionelle Untersuchungen der ABC-Transporter ABCB1
und ABCG2. PhD thesis, Rheinische Friedrich-Wilhelms Universität Bonn, Germany, 2010.
Le Goff, W., Settle, M., Greene, D. J., et al. Reevaluation of the role of the multidrug-resistant Pglycoprotein in cellular cholesterol homeostasis. J. Lipid Res. 2006, 47(1), 51-58.
Litman, T., Brangi, M., Hudson, E., et al. The multidrug-resistant phenotype associated with
overexpression of the new ABC half-transporter, MXR (ABCG2). J. Cell Sci. 2000, 113 ( Pt 11),
2011-2021.
Litman, T., Jensen, U., Hansen, A., et al. Use of peptide antibodies to probe for the mitoxantrone
resistance-associated protein MXR/BCRP/ABCP/ABCG2. Biochim. Biophys. Acta. 2002,
1565(1), 6-16.
Lua, L. H. L. and Reid, S. The importance of cholesterol for insect cell growth and baculovirus
production. Animal Cell Technology Meets Genomics. Dordrecht, Netherlands, Springer.
2005, 565-568.
Mao, Q., Conseil, G., Gupta, A., et al. Functional expression of the human breast cancer resistance
protein in Pichia pastoris. Biochem. Biophys. Res. Commun. 2004, 320(3), 730-737.
Med. Chem. Lett. 2011, 21(12), 3654-3657.
Özvegy, C., Litman, T., Szakacs, G., et al. Functional characterization of the human multidrug
transporter, ABCG2, expressed in insect cells. Biochem. Biophys. Res. Commun. 2001, 285(1),
111-117.
Özvegy, C., Varadi, A. and Sarkadi, B. Characterization of drug transport, ATP hydrolysis, and
nucleotide trapping by the human ABCG2 multidrug transporter. Modulation of substrate
specificity by a point mutation. J. Biol. Chem. 2002, 277(50), 47980-47990.
Pal, A., Mehn, D., Molnar, E., et al. Cholesterol potentiates ABCG2 activity in a heterologous
expression system: improved in vitro model to study function of human ABCG2. J. Pharmacol.
Exp. Ther. 2007, 321(3), 1085-1094.
161
Pop, N. Development of functional assays for human neuropeptide Y (Y1,2,4,5) receptors exploiting
GTPase activity and (bio)luminescence as readout. PhD thesis, University of Regensburg,
Germany, 2010.
Pozza, A., Perez-Victoria, J. M. and Di Pietro, A. Overexpression of homogeneous and active ABCG2 in
insect cells. Protein Expr. Purif. 2009, 63(2), 75-83.
Pozza, A., Perez-Victoria, J. M., Sardo, A., et al. Purification of breast cancer resistance protein ABCG2
and role of arginine-482. Cell. Mol. Life Sci. 2006, 63(16), 1912-1922.
Rendic, D., Wilson, I. B. H. and Paschinger, K. The glycosylation capacity of insect cells. Croat. Chem.
Acta 2008, 81(1), 7-21.
Robey, R. W., Medina-Perez, W. Y., Nishiyama, K., et al. Overexpression of the ATP-binding cassette
half-transporter, ABCG2 (Mxr/BCrp/ABCP1), in flavopiridol-resistant human breast cancer
cells. Clin. Cancer Res. 2001, 7(1), 145-152.
Saito, H., Hirano, H., Nakagawa, H., et al. A new strategy of high-speed screening and quantitative
structure-activity relationship analysis to evaluate human ATP-binding cassette transporter
ABCG2-drug interactions. J. Pharmacol. Exp. Ther. 2006, 317(3), 1114-1124.
Saito, H., Osumi, M., Hirano, H., et al. Technical pitfalls and improvements for high-speed screening
and QSAR analysis to predict inhibitors of the human bile salt export pump (ABCB11/BSEP).
AAPS J. 2009, 11(3), 581-589.
Sarkadi, B., Price, E. M., Boucher, R. C., et al. Expression of the human multidrug resistance cDNA in
insect cells generates a high activity drug-stimulated membrane ATPase. J. Biol. Chem. 1992,
267(7), 4854-4858.
Sarkadi, B. and Telbisz, A. ABCG2 transporter assay. WO/2013/128217 A1
(PCT/HU2013/000021), 2013.
Scala, S., Akhmed, N., Rao, U. S., et al. P-glycoprotein substrates and antagonists cluster into two
distinct groups. Mol. Pharmacol. 1997, 51(6), 1024-1033.
Senior, A. E., al-Shawi, M. K. and Urbatsch, I. L. The catalytic cycle of P-glycoprotein. FEBS. Lett. 1995,
377(3), 285-289.
Sesink, A. L., Arts, I. C., de Boer, V. C., et al. Breast cancer resistance protein (Bcrp1/Abcg2) limits net
intestinal uptake of quercetin in rats by facilitating apical efflux of glucuronides. Mol.
Pharmacol. 2005, 67(6), 1999-2006.
Telbisz, A., Muller, M., Özvegy-Laczka, C., et al. Membrane cholesterol selectively modulates the
activity of the human ABCG2 multidrug transporter. Biochim. Biophys. Acta. 2007, 1768(11),
2698-2713.
162
Chapter 6
Chapter 7
7
Summary
Multidrug transporter proteins such as ABCB1, ABCC1 and ABCG2 are known to contribute the
chemoresistance of malignant tumors. Expression of these transporters by cancer cells leads to an
efflux of anticancer drugs. The inhibition of efflux pumps, namely ABCB1 (p-glycoprotein), has been
explored in clinical trials as an approach to overcome ABC transporter mediated drug resistance in
cancer patients, but did not lead to a drug approval. The limitations of this approach can be
explained by multiple resistance mechanisms including the contribution of ABC transporters other
than ABCB1. The relevance of ABCG2 inhibition has not been adequately explored.
Apart from chemoresistance in cancer, ABC transporters play a major role as efflux pumps expressed
in epithelia and endothelial cells, influencing the bioavailability, tissue distribution, access to the
brain, and the excretion of drugs. In particular, modulation of ABC transporters at the blood brain
barrier (BBB) may considerably increase the entry of drugs into the central nervous system (CNS) and
enhance the effectiveness of pharmacotherapy, including the chemotherapy of malignancies in the
CNS, as demonstrated in a proof-of-concept study in mice with paclitaxel in combination with the
ABCB1 modulator valspodar. This strategy should even be more effective in case of inhibition of
ABCG2, as this transporter is considered to be the most important efflux transporter at the BBB. To
prove this hypothesis, potent and selective ABCG2 modulators are required.
This doctoral project was pursued in collaboration with Prof. Dr. B. König (Institute of Organic
Chemistry, University of Regensburg), aiming at the development of selective ABCG2 modulators.
Previously reported quinoline carboxamide-type compounds from our laboratory such as UR-ME22-1
were highly potent and selective ABCG2 modulators, but revealed unfavourable physico-chemical
properties, mainly due to poor solubility and rapid enzymatic cleavage of a central benzamide bond.
Next generation inhibitors, synthesized in the Institute of Organic Chemistry, were characterized in
various functional cellular assays (Hoechst 33342, calcein-AM, pheophorbide a, flow cytometric
mitoxantrone assay, kinetic cytotoxicity assay) available in our laboratory. Additional in vitro models
for the characterization of the transporter inhibitors were established (mitoxantrone microplate
164
Chapter 7
assay for ABCG2, calcein ABCC1 assay, ATPase assay for ABCG2 using Sf9 membranes), the selectivity
for ABCG2 compared to ABCB1 and ABCC1 was determined, and (bio)analytical studies (plasma
protein binding by ultrafiltration, equilibrium dialysis, isothermal titration calorimetry and HPLC;
stability in buffer and against enzymatic cleavage) were performed. Preliminary preclinical studies in
nude mice bearing orthotopically implanted genetically modified human glioblastoma enabling
optical in vivo imaging, proved that U-87MG xenografts grow rapidly in the brain. However, these
tumors were too aggressive to perform therapeutic studies. Therefore, in search for a more
appropriate tumor entity, three malignant human brain tumor cell lines (Daoy, LN-18 and SW 1783)
were characterized in vitro regarding growth kinetics, chemosensitivity against several anti-cancer
drugs, and ABC transporter expression as well as in vivo for tumorigenicity in a subcutaneous tumor
model in nude mice.
Bioisosteric replacement of the central amide core of the first and second generation of ABCG2
inhibitors by a biphenyl, quinoline, indol or a triazole moiety resulted in chemically more stable
compounds, and the introduction of triethyleneglcyol residues increased solubility. In the new
compound libraries dual ABCB1/ABCG2 and ABCG2-selective modulators were identified which are
among the most potent BCRP inhibitors reported so far.
Selected ABCG2 modulators were additionally investigated in an in vitro model of the BBB (porcine
brain capillary endothelial cell layers, expressing ABCG2; cooperation with Prof. H. J. Galla, University
of Münster) to get closer to the physiological situation and to predict the influence of the compounds
on the penetration of co-administered drugs into the brain of nude mice bearing orthotopically
growing xenografts. Surprisingly, in the BBB model the investigated inhibitors were much more
potent than in the fluorescence based cellular assays.
The reduced transport of a substrate used to determine ABCG2 activity in the presence of a
modulator may result from different modes of action: competition of the test compound with the
indicator as a substrate or inhibition of the transport mechanism. To discriminate between the
mechanisms of modulation, a colorimetric ATPase assay in the microplate format was established
using membrane preparations from baculovirus-infected Sf9 cells expressing the human ABCG2
protein. All tested compounds turned out to be ‘true’ ABCG2 inhibitors, capable of inhibiting
substrate-induced ATPase activity.
Summary
165
In summary, the new ABCG2 modulators were highly potent and selective inhibitors, but showed
poor water solubility and very high plasma protein binding (almost 100%). The drug-like properties of
these
compounds
have
to
be
further
improved
in
view
of
in
vivo
studies.
166
Chapter 7
Appendix
A
Expression of ABCG2 at the
murine blood-brain barrier immunohistochemical
investigations
A.1
Introduction
Immunohistochemical (IHC) investigations were performed to study the expression of the Abcg2
protein (murine ortholog of the human ABCG2 transporter) at the blood-brain barrier of nude NMRI
(nu/nu) mice using paraffin embedded brain tissue sections. According to an indirect method for the
detection of the transporter protein, an unlabeled primary antibody and a horseradish peroxidase
(HRP)-conjugated secondary antibody were used.
Principle of peroxidase immunohistochemical staining:
The protein of interest is targeted by an antibody that is conjugated with a peroxidase enzyme (e.g.
HRP). The antibody or target protein is visualized via a peroxidase-catalyzed reaction in the presence
of hydrogen peroxide. The substrate 3,3’-diaminobenzidine (DAB) is thereby oxidized giving a brown
precipitate:
Peroxidase + 2 H2O2
O2 + DAB
O2 + 2 H2O
insoluble, black-brown precipitate.
168
A.2
Appendix
A.2.1 Drugs and chemicals
Antibodies were obtained from Santa Cruz Biotechnology (Heidelberg, Germany) unless otherwise
stated. 0.1% PBS-T buffer was made by diluting 1000 µL Triton X-100 (Sigma, Munich, Germany) in
1000 mL of PBS.
Nuclear fast red solution (0.1% (m/v); also known as ‘Kernechtrot’) was prepared by dissolving
nuclear fast red (Sigma, Munich, Germany) in a 5% (m/v) aluminum sulfate solution. The solution was
heated (80 °C) and filtrated through a paper filter when cooled down.
All other chemicals were purchased from Merck (Darmstadt, Germany) in p.a. quality if not otherwise
stated. Millipore water was used.
A.2.2 Paraffin embedding and sectioning
In order to investigate the extent of Abcg2 transporter expression at the BBB after tumor
implantation, a nude NMRI (nu/nu) mouse bearing intracerebral U-87 MG Luc2/Katushka_Clone 15
xenografts was killed by cervical dislocation (U-87 MG Luc2/Katushka co-transfectants were obtained
as described elsewhere [Kühnle, 2010]) .
The brain was excised, fixed in Bouin´s solution (saturated picric acid/ formaldehyde/ glacial acetic
acid 15:5:1) for one day and embedded in paraffin according to a standard procedure described
elsewhere [Müller, 2007]. Serial coronal sections (6 μm) were cut every 200 μm using a Leica
RM2255 microtome (Leica; Bensheim, Germany) and mounted on microscope slides.
A.2.3 Immunoperoxidase staining
For IHC investigations, paraffin tissue sections were deparaffinized with xylene (2 x 10 min) and
rehydrated in graded isopropyl alcohol (iPrOH) followed by two washing steps (10 min) with PBS-T
buffer. Unspecific binding of the primary antibody was blocked by incubation of the tissue sections in
3% nonfat dry milk in PBS-T (blocking solution) for one hour. Thereafter, the sections were incubated
with different concentrations of the primary AB in blocking solution over night at 4 °C in Quadriperm
lux-multiplates (Greiner, Frickenhausen, Germany). Wet filter paper was added to the Quadriperm
chambers to create a humid atmosphere. To avoid wasting of the antibodies, the area around the
tissue samples was encircled with a wax pen (PAP PEN; Kisker Biotech, Steinfurt, Germany) giving a
water-repellent circle. In controls, the respective solution was added without primary AB to test the
protocol and the specificity of the antibody used.
Expression of ABCG2 at the murine BBB – immunohistochemical investigations
169
The next day, the secondary AB was added after three washing steps (10 min) with PBS-T, and
samples were stored for 1-2 hours in the dark. After additional three washing steps with PBS-T, the
tissue sections were stained via the peroxidase-diaminobenzidine reaction using a DAB Substrate Kit
for Peroxidase (Vector Laboratories; Burlingame, CA) according to the instructions for IHC staining. In
brief: immediately before use, the substrate solution was prepared by adding 2 drops of buffer stock
solution to 5 mL of distilled water. Afterwards, 4 drops of DAB stock solution and 2 drops of the
hydrogen peroxide solution were added and mixed. To obtain a gray-black stain 2 drops of nickel
solution were added to the mixture and the tissue sections were incubated with the substrate
solution at room temperature for 7-10 min. After washing in water for 5 min, nuclei were
counterstained by mounting the samples for 8 min in a 0.1% nuclear fast red solution and washed
thoroughly another three times with Millipore water. The wax pen marker was carefully removed
using xylene. After dehydration in an ascending iPrOH series and xylene, sections were mounted with
DePeX (EMS; Hatfield, PA). Stained tissue samples were analyzed with a BH-2 microscope (Olympus,
Hamburg, Germany).
Two different antibody-combinations were used for IHC:
1. primary AB:
secondary AB:
2. primary AB:
ABCG2 (BXP-53); 1:50 and 1:100
HRP-conjugated goat anti-rat IgG, 1:750
rabbit polyclonal BCRP/ABCG2 antibody (orb10178), Biorbyt Ltd.
(Cambridge,
secondary AB:
A.3
UK), 1:100 and 1:500
HRP-conjugated donkey anti-rabbit IgG, 1:1,000
Results
The visualization of Abcg2 at the BBB of a nude mouse, bearing an intracerebral U-87 MG
Luc2/Katushka_Clone 15 tumor, using the ABCG2 (BXP-53) antibody from Santa Cruz Biotechnology
(recommended for the detection of ABCG2 of mouse and human origin) was not successful. After
immunoperoxidase staining with the DAB Substrate Kit from Vector Laboratories, only erythrocytes
and blood constituents were positive, whereas Abcg2 expression was not detected at the murine
brain capillaries (Figure A.1).
170
Appendix
Figure A.1: Immunohistochemical peroxidase staining of brain tissue section of nude NMRI (nu/nu) mice bearing
intracerebral U-87 MG Luc2/Katushka_Clone 15 xenografts using ABCG2 (BXP-53) antibody from Santa Cruz Biotechnology.
A: Brain capillary with a clearly visible monocyte; B, C: Logitudinal section through a brain capillary; D: Unspecific staining of
constituents in the tumor brain area; E: Blood vessel; F: Transverse section through a murine brain capillary.
To exclude that the failure of Abcg2 detection is antibody-dependent, the same experiment was
performed with a rabbit polyclonal anti-BCRP primary antibody from Biorbyt Ltd. (Cambridge, UK),
suitable for the detection of ABCG2 of human, mouse and rat origin, in combination with a secondary
HRP-conjugated donkey anti-rabbit AB from Santa Cruz.
Expression of ABCG2 at the murine BBB – immunohistochemical investigations
171
Unfortunately, the detection of Abcg2 transporter expression in brain capillaries of nude NMRI
(nu/nu) also failed. In accordance to the immunohistochemical staining performed with the AB from
Santa Cruz, only unspecifically stained erythrocytes became visible (results not shown).
A.4
Although the presence of Abcg2 at the BBB of CF-1 mice was demonstrated previously by Real-Time
quantitative RT-PCR [Cisternino et al., 2004], the Abcg2 transporter was not detectable at the mouse
BBB of nude NMRI (nu/nu) mice by immunohistochemistry using two different antibodies, suggesting
that either Abcg2 is not expressed in the brain of nude mice or the used antibodies do not react with
the murine ABCG2 ortholog.
A.5
References
Cisternino, S., Mercier, C., Bourasset, F., et al. Expression, up-regulation, and transport activity of the
multidrug-resistance protein Abcg2 at the mouse blood-brain barrier. Cancer Res. 2004,
64(9), 3296-3301.
Müller, C. New approaches to the therapy of glioblastoma: investigations on RNA interference,
kinesin Eg5 and ABCB1/ABCG2 inhibition. PhD thesis, University of Regensburg, Germany,
2007.
Ich erkläre hiermit an Eides statt, dass ich die vorliegende Arbeit ohne unzulässige Hilfe Dritter und
ohne Benutzung anderer als der angegebenen Hilfsmittel angefertigt habe; die aus anderen Quellen
direkt oder indirekt übernommenen Daten und Konzepte sind unter Angabe des Literaturzitats
gekennzeichnet.
Weitere Personen waren an der inhaltlich-materiellen Herstellung der vorliegenden Arbeit nicht
beteiligt. Insbesondere habe ich hierfür nicht die entgeltliche Hilfe eines Promotionsberaters oder
anderer Personen in Anspruch genommen. Niemand hat von mir weder unmittelbar noch mittelbar
geldwerte Leistungen für Arbeiten erhalten, die im Zusammenhang mit dem Inhalt der vorgelegten
Dissertation stehen.
Die Arbeit wurde bisher weder im In- noch im Ausland in gleicher oder ähnlicher Form einer anderen
Prüfungsbehörde vorgelegt.
Regensburg,
..................................
Stefanie Bauer

Quinoline carboxamides as modulators of Breast Cancer Resistance

Transcription

Similar documents

INTEGRITY BATHING SYSTEM

Fadlalla, Khalda - One Medicine One Health at Tuskegee University

TechnoCompound BioBatch en

Poster - BMG Labtech

Biomedica EZ4U - Oxford Biosystems

Cell Proliferation - Bio

PDE FIZZO BROCHURE.pages

Prof. Dr. Hamdi TEMEL E-mail: [email protected] Web

ImageXpress Micro (XL) datasheet rev E.indd